Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

Mantle and geological evidence for a Late Jurassic–Cretaceous

suture spanning North America

Karin Sigloch1,† and Mitchell G. Mihalynuk2

1
Department of Earth Sciences, University of Oxford, South Parks Road, Oxford OX1 3AN, UK
2
British Columbia Geological Survey, P.O. Box Stn Prov Govt, Victoria, BC, V8W 9N3, Canada

ABSTRACT of the archipelago, whereas North America ambient mantle, and seismic waves propagate
converged on the archipelago by westward through it at slightly faster velocities. Joint sam-
Crustal blocks accreted to North America subduction of an intervening, major ocean, pling of the subsurface by thousands to millions
form two major belts that are separated by the Mezcalera-Angayucham Ocean. of crossing wave paths, generated by hundreds
a tract of collapsed Jurassic–Cretaceous The most conspicuous geologic prediction or thousands of earthquakes, enables computa-
basins extending from Alaska to Mexico. is that of an oceanic suture that must run tion of three-dimensional (3-D) tomographic
Evidence of oceanic lithosphere that once along the entire western margin of North images of the whole-mantle structure. Some
underlay these basins is rare at Earth’s sur- America.  It formed diachronously between high-velocity domains connect upward to active
face. Most of the lithosphere was subducted, ca.  155  Ma and ca.  50  Ma, analogous to subduction zones, providing a direct verification
which accounts for the general difficulty of diachro­nous suturing of southwest Pacific arcs of slab origin as cold, dense, and seismically fast
reconstructing oceanic regions from sur- to the northward-migrating Australian conti- oceanic lithosphere (e.g., for North America,
face evidence. However, this seafloor was nent today. We proceed to demonstrate that see Grand et al., 1997).
not destroyed; it remains in the mantle be- this suture prediction fits the spatio-temporal It should be possible to reconcile the sub-
neath North America and is visible to seis- evidence for the collapse of at least 11 Middle surface record of subduction, as imaged by
mic tomography, revealing configurations Jurassic to Late Cretaceous basins wedged geophysics, with the geological record of ac-
of arc-trench positions back to the breakup between the Intermontane and Insular-Guer- cretionary orogens—an integration we refer
of Pangea. The double uncertainty of where rero superterranes, about half of which are to as “tomotectonic” analysis. For Jurassic–­
trenches ran and how subducting lithosphere known to contain mantle rocks. These rela- Cretaceous North America, the prevailing in-
deformed while sinking in the mantle is sur- tively late suturing ages run counter to the terpretation of the land and marine surface rec­
mountable, owing to the presence of a spe- Middle Jurassic or older timing required and ords is that of eastward subduction of a single
cial-case slab geometry. Wall-like, linear slab asserted by the prevailing, Andean-analogue large oceanic plate, the Farallon plate, beneath
belts exceeding 10,000 km in length appear to model for the North American Cordillera. We the western continental margin since at least
trace out intra-oceanic subduction zones that show that the arguments against late sutur- 180 Ma (Hamilton, 1969; Ernst, 1970; Burch-
were stationary over tens of millions of years, ing are controvertible, and we present mul- fiel and Davis, 1972, 1975; Monger et  al.,
and beneath which lithosphere sank almost tiple lines of direct evidence for late suturing, 1972; Monger and Price, 1979). This scenario
vertically. This hypothesis sets up an absolute consistent with geophysical observations. We is analogous to today’s Nazca plate subducting
lower-mantle reference frame. Combined refer to our close integration of surface and beneath the Andean margin of South America. It
with a complete Atlantic spreading record subsurface evidence from geology and geo- has been challenged by a minority of geologists
that positions paleo–North America in this physics as “tomotectonic analysis.” This type who instead argued that Mesozoic North Amer-
reference frame, the slab geometries per- of analysis provides a stringent test for cur- ica overrode and accreted a huge archipelago
mit detailed predictions of where and when rently accepted tectonic models and offers a of intra-oceanic trenches and their arc terranes
ocean basins at the leading edge of westward- blueprint for similar, continental-scale inves- (Moores, 1970, 1998; Schweickert and Cowan,
drifting North America were subducted, how tigations in other accretionary orogens. 1975; Ingersoll and Schweickert, 1986; Inger­
intra-oceanic subduction zones were over- soll, 2008; Dickinson, 2004, 2008), which im-
ridden, and how their associated arcs and 1. INTRODUCTION plies eastward and westward subduction of more
basement terranes were sutured to the con- than one plate. The controversy was revived by
tinent. An unconventional paleogeography Subduction of ocean basins leaves two kinds Johnston (2001, 2008) and Hildebrand (2009,
is predicted in which mid- to late Mesozoic of records. At the surface, it builds accretion- 2012), who focused on westward subduction
arcs grew in a long-lived archipelago lo- ary orogens from crustal slivers that grew or beneath a ribbon continent that included peri-
cated 2000–4000 km west of Pangean North fused above subduction zones, for example, cratonic terranes.
America (while also consistent with the con- arcs, accretionary wedges, and ophiolites. In the Magnetic isochrons on the Pacific plate leave
ventional view of a continental arc in early mantle, subducted oceanic lithosphere loses its no doubt that the Farallon plate existed and
Mesozoic times). The Farallon Ocean sub- rigidity and becomes part of a viscously deform- grew to the (north-)east of the Pacific plate since
ducted beneath the outboard (western) edge ing slab that sinks toward Earth’s core under the 180+  Ma (Engebretson et  al., 1985; Atwater,
pull of gravity. A sinking slab remains visible 1989; Seton et al., 2012). Quantitative plate
†
karin​.sigloch@​earth​.ox​.ac.uk to seismic tomography because it is cooler than reconstructions (e.g., Engebretson et al., 1985;

GSA Bulletin; November/December 2017; v. 129; no. 11/12; p. 1489–1520; doi: 10.1130/B31529.1; 7 figures; 2 tables; Data Repository item 2017185;
published online 30 June 2017.

Geological Society of America Bulletin, v. 129, no. 11/12 1489

© 2017 The Authors. Gold Open Access: This paper is published under the terms of the CC-BY license.
 Sigloch and Mihalynuk

Seton et al., 2012) are based on these data, but Figure 1. Evolution of the north- A
because isochrons reflect seafloor spreading ern Pacific basin as constrained
rather than subduction, they cannot resolve the by seafloor magnetic isochron 600
ambiguity of Andean- versus archipelago-style data, modified from Enge-
paleogeography. This is illustrated by Figure 1, bretson et  al. (1985). (A)  Early
where preserved constraints on the Farallon-Pa- Cretaceous (140  Ma) recon-
cific spreading ridge are consistent with either struction of continents and oce- FAR
a margin-hugging Farallon trench (Fig. 1A) or anic plates, in a fixed hotspot
an intra-oceanic Farallon trench far outboard in (i.e., lower-mantle) reference IZA
the proto–Pacific basin (Fig. 1B), which would frame. Unfilled gray coastlines 00
have been overridden and “accreted” to the show present-day positions of
the Americas for reference. On PAC
west coast late in its life. This ambiguity was 1800 2700
clearly pointed out by early workers in marine the nascent Pacific plate (PAC),
geophysics (Engebretson et al., 1985; Atwater, west-northwest–striking iso- B
1970, 1989), but they adopted the Andean-type chrons record the existence of
trench configuration that has since prevailed in a spreading ridge and hence 600
the geologic community (Fig. 1A). of the adjacent Farallon plate

NE
The model of unchanging Farallon-beneath- (FAR). The existence of the
Izanagi (IZA) plate is similarly

GO
continent subduction has the appeal of continu-
ity and simplicity. The Farallon plate subducts recorded by ENE-striking iso- FAR

TIA
eastward into the continental Cascadia trench chrons on PAC. Blank areas of
today (Fig.  1C), and there was also eastward seafloor have since subducted. IZA

BL
subduction beneath the southwest coast of the Eastward-dipping trench (gray 00

E
United States prior to ca.  170  Ma, which built barbs) along the American west
the so-called “native Triassic–Jurassic arc” on coast marks the eastern edge of PAC
1800 2700
cratonic basement (e.g., Asmerom et al., 1990; the Farallon plate, as implied
Dickinson, 2008; Barth et  al., 2011; Saleeby by the standard “Andean-ana­
C
and Dunne, 2015), also called “Nazas arc” in logue” model. Paleolocation
Mexico (Dickinson and Lawton, 2001). By of North America (and of its 600
equating the eastward-spreading Farallon plate, subsequent westward drift) is
which existed since at least 180  Ma, with the constrained by the fully pre-
eastward subducting plate that generated the na- served isochron record of the FAR
tive arc, the Andean-analogue model interpolates central Atlantic. (B) Same as
that Farallon subduction beneath the continent in A, but with a more westerly,
would also have operated at all intermediate intra-oceanic Farallon trench PAC
times. For the Canadian segment of the Cor­di­ (black barbs). Isochron data
00
llera since ~175 m.y., this scenario was espoused are equally compatible with
by van der Heyden (1992) and many followers, this scenario. The area labeled
and recently reasserted in a review by Monger “NEGOTIABLE” remains un- 1800 2700

(2014). For the U.S. segment, Andean-style constrained by marine surface observations and hosted the “archipelago” of island arcs that
subduction has been asserted in a large body of we interpret from observed geometries of subducted seafloor. (C) Pacific basin at present day
literature (e.g., Hamilton, 1969; Burchfiel et al., (0  Ma), showing all PAC-FAR and PAC-IZA isochrons preserved on the PAC plate. Those
1992; Barth et al., 2011) for the time since birth already formed in A and B are now located around 35°N–40°N, 150°E–170°E. Slivers of active
of the Sierra Nevada batholith ca.  210  Ma, or PAC-FAR ridge and recently formed Farallon plate survive offshore the northwest United
even earlier (e.g., Saleeby and Dunne, 2015). States and Vancouver Island. Today’s west coast is located further west than the intra-oceanic
Some workers have included episodes of back- FAR trench in B, implying that the latter was overridden and “accreted” in the (relatively
arc basin development (e.g., Dickinson, 1976); recent) past to form the current, coast-hugging Cascadia subduction zone.
others have envisaged enhanced retro-arc ba-
sin development via oblique subduction (e.g.,
Saleeby, 1981, 1983; McClelland et al., 1992) happened earlier, because the vast Cretaceous nuk (2013); through tomotectonic analysis,
or eastward subduction punctuated by periods of Farallon plate in Figure  1A was devoid of ac- we found strong support for an archipelago
largely transcurrent motion (Ernst, 2011). tive arcs. The terrane accretions are naturally paleogeography. Voluminous and very deep
The challenge for Andean-style models is to accounted for by archipelago models, where slab material observed at the longitudes of the
explain a long sequence of Cretaceous arc ter- the Farallon plate was much smaller and the Farallon trench position of Figure 1B argues for
rane accretions to the west coast from Mexico “­NEGOTIABLE” region in Figure 1B was pop- intra-oceanic Farallon subduction, unless slabs
to Alaska, most notably a microcontinent com- ulated by intra-oceanic arc terranes in the style moved laterally over large and strongly variable
posed of the Insular (Peninsular-Alexander- of today’s southwest Pacific Ocean. distances after subduction. If convective dis-
Wrangellia) and Guerrero superterranes (for These conflicting scenarios were first evalu- placements in the mantle were perfectly under-
brief terrane descriptions, see Table 1). Such ated against the distribution and depth extent of stood, it would be straightforward to infer paleo­
models need to argue that these accretions subducted lithosphere by Sigloch and Mihaly­ trench locations from slab geometries alone, but

1490 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

TABLE 1. CORDILLERAN KEY COMPONENTS INCLUDING SUPERTERRANES, MAGMATIC BELTS AND INDIVIDUAL TERRANES, AND
THEIR SLAB/BASIN AFFILIATIONS FOR EARLY JURASSIC AND LATER TIMES (UNTIL AMALGAMATION WITH NORTH AMERICA)
Relation to geological Geophysically affiliated Mesozoic
Description of constituent components superterranes SLAB and ocean basin
Cordilleran superterranes

Alaskan Cretaceous arc Koyukuk, Nyak, and Togiak terranes. ANG slab and Angayucham
terranes Ocean.
Insular superterrane Wrangellia (WR), Alexander (AX) and Peninsular (PE) arc terranes in Canada Northern MEZ slab and northern
(INS) and Alaska. Mezcalera Ocean.
Intermontane In Canada, Quesnel (QN) and Stikine (ST) arc terranes, and Cache Creek Cache Creek Ocean, a precursor
superterrane (IMS) Ocean terrane (CC); in conterminous U.S., terranes of Blue Mountains (BM); of Mezcalera Ocean. No slab
farther south, Native Triassic-Jurassic arc (NJ), which includes Triassic- imaged (yet).
Jurassic episode of the Sierra Nevada batholith; in Mexico, Nazas arc, the
southward continuation of NJ arc.
Guerrero superterrane Considerations range from an intraoceanic arc complex united by Mesozoic Southern MEZ slab and
(GUS) of Mexico volcanic crust (Dickinson and Lawton, 2001; which includes Guerrero, southern Mezcalera Ocean.
Arteaga, Papanoa, Santa Ana, and upper plate components of Vizcaino
and Magdalena terranes of Silberling et al., 1992) to a composite of various
terranes related to a long-lived Andean-type arc (e.g. Tahue, Arcelia,
Zihuatanejo, and Guanajuato terranes of Centeno-Garcia et al., 2011).

Cordilleran magmatic belts

Coast Cascades Orogen 1600-km-long belt extending from Yukon to Washington, cored by 105-45 Ma Formed by eastward Farallon
plutonic and metamorphic rock (Monger, 2014). Overprints older parts of the subduction after override of
Mezcalera Ocean suture between IMS and INS. INS arcs.
Cretaceous Sierra Cretaceous magmatic pulse (ca. 125–85 Ma) overprints the suture of Mezcalera Formed by eastward Farallon
Nevada batholith of Ocean between older Native arc (IMS) and recently accreted INS. subduction, after override of
California (SN) INS arcs.
Peninsular batholith of Cretaceous magmatic pulse (ca. 125–85 Ma) in north overprints the suture of Formed by eastward Farallon
Mexico Mezcalera Ocean between older Triassic-Jurassic arc (IMS) and recently subduction, after override of
accreted GUS. Zircon ages on the western zone (Silver and Chappell, 1988) GUS arcs.
range from 140 to 105 Ma, whereas those of the easternmost side of the
batholith are much younger at 85–75 Ma (Grove et al., 2003).

Cordilleran terranes

Alexander terrane (AX) Metamorphosed Neoproterozoic arc (Gehrels et al., 1996); Cambro-Ordovician Part of INS. Stationed above MEZ slab in
arc (Beranek et al., 2012); Ordovician to Triassic marine quartzo-feldspathic Mesozoic, underridden by
clastic and carbonate shelf, deep water off-shelf and Late Triassic rift Mezcalera Ocean.
assemblages; pinned to Wrangellia by ca. 308 Ma (Gardner et al., 1988).
Angayucham terrane Imbricated Middle Devonian to Jurassic oceanic crustal succession including Part of the Alaskan Early Angayucham Ocean floor.
pelagic strata, greywacke and limestone. arc complex.
Bridge River terrane Mississippian to Jurassic oceanic crustal succession, disrupted within Part of INS. Mezcalera Ocean and its
(BR) accretionary complex, includes Late Triassic blueschist. precursor.
Cache Creek terrane Mississippian to Early (and perhaps Middle) Jurassic oceanic assemblage Part of IMS. Cache Creek Ocean, a precursor
(CC) including mantle tectonite, Mid Permian gabbro (Mihalynuk et al., 2003), of Mezcalera Ocean.
supra-subduction zone basalt (Ash, 1994), radiolarian chert, primitive arc
basalt, ocean island basalt and carbonate platform succession containing
exotic Tethyan faunas of Middle Permian (Monger and Ross, 1971) to Middle
Triassic age (Orchard et al., 2001); blueschist of Late Triassic (Paterson and
Harakal, 1974) and Middle Jurassic age (174 Ma, Mihalynuk et. al, 2004).
Chugach terrane (CH) Turbiditic wacke and argillite assemblages of Late Jurassic to Early Cretaceous, Outboard of Offscrapings of northern Farallon
mid Cretaceous and Late Cretaceous ages (Amato et al., 2013); interpreted Cretaceous INS. Ocean/Cascadia Root slabs
as an accretionary complex (Berg et al., 1972). Inner (older) parts include CR and CR2.
Triassic to Cretaceous chert and Permian limestone with Tethyan faunas and
205–192 Ma blueschist (Roeske et al., 1989; López-Carmona et al., 2011).
Franciscan composite Cretaceous and Tertiary accretionary complex including many fault-bounded Outboard of Offscrapings of southern
terrane (FR) subterranes. Maximum depositional age for inboard (oldest) portion is ca. Cretaceous INS. Farallon Ocean/southern
123 Ma (Dumitru et al., 2010). Farallon slabs.
Guerrero (GU) and Late Jurassic to Early Cretaceous andesitic volcanic rocks and coeval marine to Part of GUS. Arc complex stationed
Santa Ana (SA) non-marine sedimentary rocks. above southern MEZ slab/
terranes of Mexico underridden by Mezcalera
Ocean.
Mystic (Farewell) terrane Strongly deformed Ordovician to Permian argillite, chert, sandstone and Substrate of Alaskan/ ANG slab/Angayucham Ocean.
of Alaska limestone; undated mélange; Ordovician and Triassic pillow basalt; Devonian ANG arcs.
and Triassic black shale and phosphorite; faunal ties with Siberia (Bradley
et al., 2007).
Pacific Rim terrane (PR) Late Jurassic to Early Cretaceous, tectonized sedimentary and volcanic rocks, Outboard of INS. Farallon Ocean/CR slab.
and Late Triassic limestone; Cretaceous high P/T metamorphism (Brandon,
1989); final emplacement and cooling between Early and Mid-Eocene
(Groome et al., 2003) during underthrusting of Siletz-Crescent terrane (SC).
(continued )

Geological Society of America Bulletin, v. 129, no. 11/12 1491

 Sigloch and Mihalynuk

TABLE 1. CORDILLERAN KEY COMPONENTS INCLUDING SUPERTERRANES, MAGMATIC BELTS AND INDIVIDUAL TERRANES, AND
THEIR SLAB/BASIN AFFILIATIONS FOR EARLY JURASSIC AND LATER TIMES (UNTIL AMALGAMATION WITH NORTH AMERICA) (continued)
Relation to geological Geophysically affiliated Mesozoic
Description of constituent components superterranes SLAB and ocean basin
Cordilleran terranes (continued )

Peninsular terrane in Permian limestone, Late Triassic limestone, argillite and basalt, Lower Jurassic Part of INS. Stationed above northern MEZ
Alaska (PE) volcaniclastic rocks and coeval plutons, Middle Jurassic to Cretaceous marine slab, underridden by northern
sedimentary rocks. Mezcalera Ocean.
Quesnel terrane (QN) Late Triassic and Early Jurassic arcs constructed on relicts of Silurian to Part of IMS. Built by subduction of Cache
Mid Triassic arc volcanic and marine sedimentary strata with oceanic to Creek Ocean, a precursor of
pericratonic substrate. Mezcalera Ocean.
San Juan composite Multiple nappes variably composed of Paleozoic to Mesozoic argillite-chert, Forms part of Mezcalera Ocean?
terrane basalt, ultramafite, limestone with Tethyan faunas in mélange; blueschist of INS-IMS suture.
Permian, Late Jurassic and Early Cretaceous age; youngest rocks in nappes
are 114 Ma and overlying sediments are ca. 50 Ma (Brown, 2012).
Siletz-Crescent terrane Eocene tholeiitic pillow basalt, breccia and subaerial flows; lower parts include Outboard of Farallon Ocean crustal welt/
(SC) sheeted dykes and gabbro of a partial oceanic crustal section; upper parts Cretaceous INS. CR slab.
intercalated with continentally derived sediment (Massey, 1986); aged
46–56 Ma (references in McCrory and Wilson, 2013).
Stikine terrane (ST) Late Triassic to early Middle Jurassic arc constructed on Early to mid-Triassic Part of IMS. Built by subduction of Cache
chert, widespread Permian limestone and Early Devonian to Permian arc with Creek Ocean, a precursor of
oceanic to pericratonic substrate (Logan et al., 2000). Mezcalera Ocean.
Vizcaino terrane (VC) Triassic to Cretaceous oceanic crustal and arc strata and extension of Outboard edge Farallon Ocean and offscrapings.
Franciscan accretionary complex. of GUS.
Western Jurassic/ Late Triassic to Late Jurassic arc, oceanic crustal and volcaniclastic strata Part of INS. Products of Mezcalera Ocean
Foothills composite includes numerous terranes in Klamaths and western Sierra Nevada subduction.
terrane (WF) metamorphic province that are west of the Calaveras terrane/belt as per
Dickinson (2008) and Schweickert (2015).
Wrangellia composite Devono-Mississippian arc, Pennsylvanian-Permian volcanic and marine strata Part of INS. Stationed above northern
terrane (WR) including widespread limestone and Late Permian to Mid Triassic chert, up to MEZ slab (after Triassic),
~6 km of intraplate ca. 230–225 Ma submarine to subaerial basaltic plateau underridden by Mezcalera
flows capped by Late Triassic deep water limestone and spiculitic argillaceous Ocean.
strata (Greene et al., 2010). Interbedded and overlying Late Triassic to Middle
Jurassic volcanic and clastic rocks and coeval plutons (Nixon and Orr, 2007).
Yukon-Tanana Pericratonic strata and Paleozoic to early Mesozoic arc (Stikine/Quesnel Part of IMS. Ophiolitic components may
composite terrane equivalents), lesser oceanic crust, polydeformed and metamorphosed to be old Angayucham basin,
(YTT) amphibolite and rare eclogite facies; terrane-specific cooling ca. 200–190 Ma associated with a precursor of
(Mortensen and Jilson, 1985; Newberry et al., 1998; Dusel-Bacon et al., 2002; ANG slab.
Knight et al., 2013; Staples et al., 2013).
Note: Terranes or geological belts composed of Jurassic–Cretaceous basinal strata trapped within the INS-IMS suture are described within the text. Descriptions are
after Silberling et al. (1992) unless otherwise noted. Acronyms for terranes are defined only if they are used in Fig. 4 or in the main text.

this is stymied by large uncertainties about man- other than just vertically down under stationary Section 2 (geophysical observations) and
tle rheology. The archipelago model implies trenches. Slab walls thus appear to directly map Section 3 (geological observations) act as two
tight spatio-temporal correlation between slab out paleotrench positions. Guided by the verti- complementary halves that form the core of this
and paleotrench geometries, and thus a unifor- cal sinking hypothesis, volcanic arc terranes can study. Sigloch and Mihalynuk (2013) laid out
mity of slab sinking (rheology) that is geophysi- be positioned in a mantle reference frame be- the premises for testing geologic predictions
cally appealing and a fundamental constraint cause they must have grown above the subduc- made by geophysics, and started this testing by
for mantle convection models. In contrast, the tion zones feeding the slab walls; i.e., surviving arguing that the orogenic deformation record
Andean-analogue model implies weak correla- tectonic blocks can be paleopositioned without of the Cretaceous Cordillera is more consistent
tion of slab to paleotrench geometries and hence reference to geologic observations. The geologic with archipelago paleogeography than with the
complicated sinking (limited predictive power land record then remains as an independent data prevailing Andean analogue. Here, we focus on
on rheology), but the trench lines would have set for testing the predictions made by geo- geologic evidence for the starkest prediction of
been relatively simple (a single, margin-hugging physical observations and the vertical sinking the archipelago model: the existence of a m­ ajor
Farallon trench). The apparent simplicity of the hypothesis, i.e., that terrane collisions occurred Jurassic–Cretaceous ocean in addition to the
Andean model presumably accounts for much when the drifting continental margin (as recon- Farallon Ocean, and its diachronous closure
of its appeal over the past decades. Contribut- structed, e.g., by Seton et  al., 2012) started to by westward subduction under the archipelago,
ing factors have been the non-consideration of laterally overlap the imaged slab walls (e.g., ahead of North America riding into its arc ter-
subsurface observations and the limitations of Sigloch, 2011). Such absolute paleoposition- ranes. The implied Late Jurassic–Cretaceous
geologic observations in challenging terrain. ing constraints on where, when, and how North suture, which should span the entire western
Investigation of the archipelago model is fa- America overrode the intra-oceanic arcs are not North American margin, is imminently testable
cilitated by the presence of a special-case slab available for archipelago models based purely on against the land record. The spatial scope of
geometry under North America: very steep land geology (Moores, 1970, 1998; Schweickert this investigation extends back to Early Jurassic
and voluminous belts of slab “walls,” which and Cowan, 1975; Ingersoll and Schweickert, (201–174 Ma, based on the time scale of ­Cohen
are difficult to explain by any sinking process 1986; Ingersoll, 2008; Dickinson, 2004, 2008). et  al. [2013], which is used throughout), as

1492 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

compared to mainly Cretaceous units in Sigloch Ultimately our study is centered in observa- thanks to the continent-spanning USArray
and Mihalynuk (2013). tions, which we argue are sufficient to falsify seismological experiment (e.g., Burdick et  al.,
The discussion of the premises for geologic the Andean-analogue model for North America, 2008; Sigloch et  al., 2008; Schmandt and
hypothesis testing is also extended in several re- because observations directly contradict a Far- Humphreys, 2010; reviewed and compared by
spects. We give a detailed accounting of Farallon allon trench continuously hugging the ancestral Pavlis et al., 2012).
subduction because lack of this discussion made continental margin from 180+ Ma to present. Here, we interpret the finite-frequency inver-
it difficult for geophysicists to engage with archi- sion of broadband, teleseismic P waveforms
pelago models based in geology. We also lay out 2. SUBDUCTED SLABS CONSTRAIN by Sigloch (2011). This study resolves deeper
the close tectonic analogy between diachronous CORDILLERAN TECTONICS structure than most, because, alongside the dense
archipelago override by Mesozoic North Amer- array data from the United States, it incorporates
ica, and override of the southwest Pacific archi- This section describes geophysical observa- data from most North American broadband sta-
pelago by present-day Australia. This makes the tions and the predictions (or hindcasts) they tions deployed since 1999. Finite-frequency
geologic predictions of section 3 more tangible, make about paleogeography, to be tested against modeling of realistic wave sensitivities on a
especially the generalized consequences of dia- the geological observations in section 3. Steep, global grid adds resolution at depth, compared
chronous ocean basin suturing at both flanks of wall-like slabs stand out among the geometries to conventional, ray theoretical imaging (Dahlen
northern Australia and expectations of analogous of subducted lithosphere surveyed in section et al., 2000; Sigloch, 2008). Figure 2A is a three-
suturing recorded by Cordilleran geology. 2.1 This geometry suggests almost vertical slab dimensional bird’s-eye rendering of the North
Section 4 delineates our findings from sinking relative to the lower mantle as the for- American slabs, showing sublithospheric areas
archi­pelago models based purely on geology mation mechanism, combined with deposition of faster-than-average P-wave velocities relative
(Moores, 1970, 1998; Schweickert and Cowan, beneath stationary trenches (section 2.2), which to the spherically symmetric mantle reference
1975; Inger­soll and Schweickert, 1986; Inger­ could only have been intra-oceanic. Massive model IASP91 of Kennett and Engdahl (1991).
soll, 2008; Dickinson, 2004, 2008) and dis- slab volumes suggest long-lived and hence old As a first-order observation, the lower-mantle
tinguishes it from ribbon continent models (i.e., Jurassic) subduction (section 2.3), which slab assemblage in Figure 2A is strikingly seg-
(Chamber­lain and Lambert, 1985; Lambert again implies intra-oceanic trenches, because mented, and it reaches equally deep in the east
and Chamberlain, 1988; Johnston, 2001, 2008; the North American margin lay further east at and in the west. Steep, wall-like slabs run in two
Hilde­brand, 2009, 2012, 2015). that time. After matching the magnetic isochron linear belts totaling >10,000 km in length. The
Section 5 raises discussion points, such as record of the Farallon plate to subducted litho- Angayucham slab wall (ANG) strikes north-
oblique collisions as the default regime; the sphere (section 2.4), much slab remains unac- west from present-day Nova Scotia to Yukon,
relevance of geodynamic convection modeling; counted for. It must represent an additional, and the Mezcalera slab wall (MEZ) runs from
prior engagement of geophysicists with (only) major ocean (section 2.5), which again implies Nova Scotia south beneath the eastern seaboard
the Andean-analogue model; slab sinking rates; intra-oceanic subduction, of both it and the Far- and the Caribbean. In depth extent, both belts
and the current limitations of our approach in allon Ocean from opposite sides, i.e., an archi- fill the middle third of the mantle from ~800 to
revealing paleogeography older than Early Ju- pelago geography. Certain eastward-dipping 2000+ km deep. An equally deep slab is found
rassic. Mantle convection modeling has not en- slab geometries are consistent with this hypoth- further west, labeled “CR” for Cascadia root,
gaged with either the archipelago model (ours esis, rather than supporting Andean-style sub- which extends from the lower mantle to the ac-
or earlier versions) or ribbon continent scenarios duction, as previously interpreted (section 2.6). tive Cascadia subduction zone at the surface.
and so has not tested them nor weighed their These slabs are ideal for estimating slab sinking (For better viewing of this western subduction,
merits relative to the Andean-analogue model. rates. Section 2.7 spells out the predicted se- the eastern slab above 800 km is not rendered
The Andean analogue has been investigated and quence of archipelago override by North Amer- in the foreground of Figure 2A, but can be in-
found to require complex mantle rheologies and ica, a scenario that may seem complex but is in spected in Supplementary Movie M1.1 Figure 3
sinking behaviors in order to deposit slabs that fact closely analogous to the override of today’s provides an oblique 3-D rendering of only Cas-
resemble the observed, lower-mantle assem- southwest Pacific archipelagos by Australia cadia/CR subduction.
blage under North America (e.g., Bunge and (section 2.8), predicting similar observable geo- Figure  4A shows the same slab assemblage
Grand, 2000; Liu et al., 2008). Our archipelago logical consequences. as Figure 2, but in map view and rendered only
scenario would produce the observed slabs with- at depths below 1100  km. MEZ and ANG are
out needing to invoke such complexities, because 2.1. Subducted Lithosphere Under North seen to be paralleled by several shorter and
the near-vertical sinking it implies has been ob- America and the Problem of Provenance more scattered, but equally deep fragments
served in some numerical convection simula- further west (CR, CR2, SF1). (We write slab
tions (e.g., Steinberger et al., 2012). Newer The mantle under North America has long names such as MEZ and ANG in uppercase to
studies of viscous slab folding (e.g., Ribe et al., been known to hold one of the largest accu- conceptually distinguish them from their associ-
2007; Stegman et al., 2010; Gibert et al., 2012; mulations of subducted lithosphere observed ated paleo-­oceans and arc terranes, e.g., Mezca­
Čížková and Bina, 2013) lend further credence anywhere. Revealed by early regional-scale to- lera and Angayucham.) The name “Mezcalera
to near-vertical sinking because they reveal a mographies (Grand, 1994; Grand et al., 1997; Ocean” is adopted from Dickinson and Law-
formation mechanism for the wide, voluminous van der Hilst et  al., 1997), this finding has ton’s (2001) review of Mexican arc geology.
slab walls that are central to our argument for been confirmed by all global tomography stud- Angayucham Ocean is named after an ophiolitic
old, intra-oceanic subduction. (Note that we use ies since (e.g., Montelli et  al., 2006; Li et  al., terrane in Alaska (Coney et al., 1980, p. 197).
the following conventions: Slab “height” refers 2008; van der Meer et al., 2010; Ritsema et al.,
to the slab’s vertical dimension, slab “length” is 2011; Obayashi et  al., 2013). Imaging resolu- 1
GSA Data Repository item 2017185, Movie M1, is
the trench-parallel dimension, and slab “width” tion of subducted slabs beneath North America available at http://​www​.geosociety​.org​/datarepository​
is the trench perpendicular dimension.) has improved enormously over the past decade /2017 or by request to editing@​geosociety​.org.

Geological Society of America Bulletin, v. 129, no. 11/12 1493

 Sigloch and Mihalynuk

Effective use of slabs for paleogeographic est (lower) parts of the MEZ and ANG slabs had 2.2. Null Hypothesis of Vertical Slab
reconstruction requires the validation of a key subducted by the time Pangea started to break Sinking in the Lower Mantle
postulate: Slab geometries preserve paleo- up, then they were deposited west of western
ocean and paleotrench geometries. Although the North America beneath intra-oceanic arcs (the These paleopositioning arguments can be
youngest (shallowest) ends of some slabs, such archipelago) and were fed by westward subduc- quantified. Figures 4A and 4B draw trench lo-
as CR, can be traced to present-day subduction tion of the seafloor shaded cyan in Figures 4A cations vertically above the slab walls, illustrat-
zones, older and deeper parts of slabs are not and 3B. If they subducted later, while the west ing our null hypothesis that every parcel of slab
dateable per se, unless they can be confidently coast was already traversing the current slab wall entered the mantle above its present-day lo-
linked to a well-constrained geologic record. resting area, then their trenches were proxi- cation and simply sank downward; i.e., it is still
If lower-mantle slabs are heavily deformed by mal to the continental margin and presumably located at the same longitude and latitude where
mantle currents, paleotrench locations could right along it (the Andean-type scenario). The it entered its trench, analogous to hotspots not
be difficult to reconstruct. The rheological latter leaves the challenge of explaining how moving significantly relative to the lower mantle
properties of slab and ambient mantle are un- a (Farallon) trench that migrated continuously (Morgan, 1981). To the extent that this vertical
constrained enough that deformation styles and smoothly with the west coast generated the sinking hypothesis is correct, and to the extent
and sinking rates have remained uncertain. It is steep, linear slab belts that bear no resemblance that the plate reconstruction is correct, Figure 4
debatable how much lateral displacement any to the shape of the continental margin and are gives absolute paleodistances of the continent
given parcel of slab might have undergone while separated by slab-free zones (an issue revisited to the trenches and their attendant arc terranes.
sinking in the mantle, i.e., how much its current in section 5.3). Our main purpose, however, is to The cumulative observational uncertainties of
latitude and longitude differ from its  trench argue the opposite scenario, i.e., of early, intra- tomographic model, reconstruction, and conti-
entry point. How can we then approach the
­ oceanic subduction. nental margin shape over time are a few ­hundred
inference of paleotrench locations and timing
from slab geometries?
Figure  4A superimposes on the slabs the
position of North America ca.  110  Ma, and Figure 2 (on following page). (A) Seismically fast domains in the lower mantle beneath North
Figure  4B shows North America for 170  Ma, America, according to the tomographic P-velocity model of Sigloch (2011). Isovelocity con-
140 Ma, and present-day. Taken together, they tours, with color changing every 200 km in depth, enclose mantle regions where wave veloci-
trace out the westward migration of the con- ties are dVp/Vp > 0.35% faster than average. These regions are interpreted as subducted,
tinent and its west coast relative to the lower cool lithosphere that was deposited beneath volcanic arcs and accumulated to form slabs.
mantle. (Movie M1 shows more detail, in 5 m.y. Three-dimensional (3-D) image is an oblique elevation view from the east-northeast. Spheri-
increments [see footnote 1].) The continent’s cal geometries have been flattened so as to minimize horizontal or vertical distortion at
drift relative to Africa and Europe is constrained the center of the rendering volume. Slabs names are capitalized acronyms that allude to
by magnetic seafloor isochrons from the fully each slab’s interpreted plate origins: MEZ slab consisting of Mezcalera Ocean lithosphere;
preserved Atlantic spreading record, with uncer- ANG—Angayucham; CR and CR2—Cascadia Root, i.e., northern Farallon Ocean; SF1—
tainties of generally only ~1° or <2 m.y. (Müller Southern Farallon; K—Kula. The 10,000-km-long and near-vertical MEZ/ANG slab walls
et al., 2008). in the lower mantle (formerly presumed Farallon slabs) are seen in the foreground (east
A lower-mantle reference frame is the natural and north). In the background (west), the lower-mantle CR is upward-continuous to to-
choice for this surface reconstruction because day’s Farallon/Juan de Fuca Trench in the Cascadia subduction zone (shallowest, purple
the slab walls are located in the lower mantle, material). The MEZ/ANG slabs have been masked above 800 km to leave an unobstructed
and our goal is to establish their paleotrench view of this entire Farallon subduction system; note the near verticality of the CR slab
locations relative to the continent’s west coast. in the lower mantle. The interpreted Mesozoic Mezcalera-Angayucham Oceans were lo-
The best observational proxy for a lower-mantle cated at the surface to the east and north of the MEZ/ANG slab walls, i.e., closest to the
reference frame is given by volcanic hotspot viewer. (B)  Cartoons show interpreted generation of the slabs; they are drawn to scale.
tracks (Morgan, 1981; Duncan and Richards, Panels B1 (Early Cretaceous) and B2 (today) interpret slab and surface evolution along
1991), and Figure  4 uses an Indo-Atlantic southerly cross section X-X′-X″ of panel A. Panels C1 and C2 interpret slab evolution under
hotspot frame back to 100 Ma (O’Neill et  al., northerly cross-section Y-Y′-Y″. (B1) MEZ slab wall was deposited by Jurassic–Cretaceous
2005) and a hybrid paleomagnetism frame for westward subduction of the Mezcalera Ocean into a stationary, intra-oceanic trench and
earlier times (Steinberger and Torsvik, 2008), beneath the Insular superterrane (INS), until the Mezcalera Ocean was consumed and the
subject to uncertainties of less than 5°. Even Insular superterrane had accreted. NJ arc is the older Native Jurassic arc intruded in stable
with additional uncertainties considered, for ex- North America. (B2) Subsequently, the southern Farallon Ocean subducted eastward be-
ample, on the shape of North America’s paleo­ neath a migrating, continental trench, depositing the eastward-dipping slab L1 (not ren-
margin, all reconstructions of this type agree dered above 800 km in A but visible in Movie M1 [see text footnote 1] and contoured green in
that Pangean North America (reconstruction Fig. 4A). MEZ slab continued to sink in place. (C1) Double-sided, intra-oceanic subduction
for 170 Ma) was located well east of the most of Angayucham and northern Farallon oceans in Jurassic–Cretaceous times generated the
easterly slab walls MEZ and ANG in Figure 4 ANG and CR slab walls. Consumption of the Angayucham Ocean ended with accretion of
(cf. Sigloch and Mihaly­nuk, 2013; GSA Data the Alaskan arcs to the continental margin pre-modified by earlier accretion of the Inter-
Repository material [see footnote 1]). montane superterrane (IMS). Later, the Farallon trench converted from intra-oceanic to
Assuming the lower-mantle slabs did not continental upon override, and proceeded to deposit a laterally sprawling, upper-mantle
­enter the mantle thousands of kilometers east of slab (blue-purple shades in panels C2 and A). Observed thickening of slabs is attributed to
their current location, then two alternatives for folding in the transition zone. All slabs sink vertically relative to the lower mantle; station-
their formation must be entertained: If the old- ary trenches produce vertical slab walls; migrating trenches produce dipping slabs.

1494 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

A X (west) Gorda Y

X′

CR Y′

X′′
(east)

b
Mezca

la
lera Y′′
0 km

S
Sla am
b g a y uch 670 km

An 1000 km

1800 km

B1 C1
X′ X′′ Y Y′ Y′′
Mezcalera Angayucham
Early Cretaceous

S. Farallon Ocean N. Farallon Ocean Ocean

0 km
N. Am. Alaskan N. Am.
INS
arcs
670 km
1000 km
NJ arc IMS
MEZ Cascadia ANG
Root (CR)

2000 km

B2 C2
X X′′ Y Y′′

Pacific North America Atlantic North America Atlantic

Gorda
Today

670 km

L1

CR
ANG
MEZ

Figure 2.

Geological Society of America Bulletin, v. 129, no. 11/12 1495

 Sigloch and Mihalynuk

Yellowstone 53° N
hotspot
N

98
Mendocino

.5°
E
triple junction

0 km
28° N
12
6°
E

670 km

1000 km
1000 km

2000 km

Figure 3. Farallon subduction beneath the Cascadia subduction zone, from trench to ~1500 km depth. Three-dimensional, oblique elevation
view from the southwest; topography and bathymetry of the western U.S. are shown as translucent surface (elevations strongly exagger-
ated; same surface is also shown schematically translated to 1000 km depth for spatial reference). Seismically fast domains according to
the tomographic P-velocity model of Sigloch (2011) are iso-contoured in 3-D where wave velocities are dVp/Vp > 0.35% faster than average
(as in Figs. 2 and 4, and using same rainbow color scale to indicate depth). Fast velocities that are clearly separate from this continuous
system are masked out, including cratonic root and transition zone slab to the east, and lower-mantle slabs CR2 and K to the north and
west (the latter are visible in Figs. 2 and 4 and Movie M1 [see footnote 1]). The slab dips eastward, but extends not farther east than the
Rocky Mountain front (see also Figs. 4A/B). The slab’s lower-mantle parts, termed “CR” or “Cascadia Root” in the text, are much more
massive than in the uppermost mantle; slab thickening happens mostly in the transition zone (light blue level, 400–600 km). Comparison
to the different perspectives of Figures 2A and 4A reveals a rotation in deposition direction. The base of CR at the red/yellow levels strikes
NW-SE (almost perpendicular to this viewing angle) and is built steep and wall-like. From the blue-green level (~800 km) up, the slab is less
massive, heavily fragmented, and shows a clear slope toward the trench, i.e., westward-shallowing; the strike of this material is more N-S,
c.f. Figure 2A. (We interpret this upper-mantle slab as deposited after CR trench had accreted and while it was dragged westward as an
Andean-style trench, i.e., post-ca. 60 Ma. This matches a clockwise rotation of conjugate magnetic isochrons on the Pacific plate from NW-
SE to more N-S, recording the fracturing of the Vancouver fragment from the northern Farallon ca. 52 Ma, as discussed in Sigloch, 2011.)

k­ilometers, or 3–24  m.y., when converted to with their arc terranes, and the oceans that inter- netized seafloor, the hotspot reference frame) to
temporal uncertainties via the drift rate of North vened between the west coast and the trenches build the collision model, is testable against the
America (for this uncertainty analysis, see sup- would have closed (the Mezcalera and Angayu- geologic record of arc terrane collisions along
plement in Sigloch and Mihalynuk, 2013). cham Oceans). This prediction, which follows the margin, which constitutes a completely
At the times and locations that North Ameri- from the null hypothesis of vertical sinking, inde­pendent set of observations. Sigloch and
ca’s west coast in Figure 4A starts to overlap the and which uses only geophysical observations Mihaly­nuk (2013) tested it against the Creta-
MEZ and ANG slabs, it is predicted to collide (tomography, plate reconstructions from mag- ceous record of marginal deformation in time

1496 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

and space, and here we show that the predicted examination of this idea on actual slab obser- down piles of periodic folds, as cartooned in
closure of the Jurassic–Cretaceous Mezca­lera vations. Southern Farallon slab L1 is another Figures 2B and 2C, where individual folds are
and Angayucham Oceans can also be demon- dipping slab deposited by a migrating, conti- not (yet) resolvable by tomography. This fold-
strated in the geologic record. Hence, we ­argue nental trench, as discussed in the caption of ing mechanism for viscous thin sheets against
that the vertical slab sinking hypothesis is hold- Figure 2. a backstop has been revealed by recent geo-
ing up to scrutiny. dynamic modeling studies (Bellahsen et  al.,
2.3. How Much Lithosphere Is in the Slabs?
The potential for demonstrating a new 2005; Ribe et al., 2007; Funiciello et al., 2003;
lower-mantle reference frame, equivalent to the The near-vertical MEZ and ANG slab walls Stegman et al., 2010; Čížková and Bina, 2013;
hotspot frame but reaching further back in time, in the foreground of Figure  2A fill more than Garel et al., 2014; see Discussion, section 5.2).
provides the strongest motivation for vetting the 1000 km of the mantle column, from ~800 km A folding process that lays down a
vertical sinking hypothesis. Its simplicity re- to >1800 km in depth, and run over 10,000 km 400–700-km-wide slab wall can be expected to
quires no input of uncertain mantle rheologies in long in an angled, highly structured geometry. generate a few hundred kilometers of deviation
order to build a subduction model, and it permits In this depth range, the CR slab is also near ver- from strictly vertical slab descent, unless trench
rigorous integration of geological observations tical; it only slopes toward its Cascadia trench motion exactly tracks the oscillatory slab fold-
into what has been a largely geophysical debate. in the upper ~800  km (most clearly seen in ing motion. It is unclear whether to picture the
Even if more complicated slab sinking were Fig. 3). In their narrowest dimension, the MEZ, trench as firmly centered above the slab wall, or
expected a priori, falsification of this simplest ANG, and CR slab walls are 400–700 km wide, always off to one side, or oscillating back and
possible sinking mode should be attempted first. best seen in Figure  4A. If each represented a forth as folds are laid down. Hence, the concept
This cannot happen so long as intra-oceanic single sheet of lithosphere dipping steeply into of “vertical sinking” of thickened lower-mantle
trenches are not considered to be an option. Ear- the lower mantle, then their width should equal slabs is meaningful only within observational
lier work has only shown the incompatibility of typical lithospheric thickness, on the order of uncertainties of a few hundred kilometers
vertical sinking with an Andean-style Farallon 100 km. Widening of steep lower-mantle slabs (roughly the half-width of the slab wall), but
trench (cf. section 5.3). under North America is a robust, consistent ob- this is still narrow compared to lower-mantle
Vertical sinking of massive slabs is plausible servation that was evident in the earliest tomo­ features or plate dimensions at the surface.
a priori because gravity is the only primary graphic images (Grand, 1994) and has been re-
2.4. Which Slabs Are Farallon Lithosphere?
force acting on them. Nonvertical displacement produced consistently by all later studies. This
of slabs would require their entrainment in the slab widening might not have been interpreted Of the lower-mantle slabs in Figure 2A, only
lateral flow of ambient mantle (“mantle wind”), due to lingering doubts about artificial image CR connects upward to the currently active
which is driven by lateral pressure gradients. In smearing, but with modern waveform imag- subduction zone of the Farallon plate (Juan de
order to be effective, the driving agents would ing methods and much denser instrumentation, Fuca and Gorda) and can hence be reliably con-
need to be even larger slabs (or plumes), which these doubts have been settled (see Discussion, sidered Farallon lithosphere (see also Fig.  3).
do not exist nearby. The North American slab section 5.2). Provenance of the remaining slabs is a priori
walls are among the most massive in the en- Assuming mature lithospheric thickness of questionable, but given that the Farallon Ocean
tire mantle and should therefore sink more 100  km, a 1000-km-“high” and 400–700-km- undoubtedly existed and spread since 180+ Ma
vertically than anything else. Also, the lower- wide slab wall would contain a paleo-ocean in (Engebretson et  al., 1985; Atwater, 1989),
mantle reference frame empirically “works” for which at least 4000–7000  km of lithosphere it is not warranted to hypothesize additional
plumes (Morgan, 1981; O’Neill et  al. 2005), would have been generated, and even more if oceans until all inferred Farallon seafloor has
confirming sluggish lateral convection at those the subducted lithosphere was young and pro- been matched to appropriate slabs. Farallon
depths. It may therefore work equally well, or portionally thinner. It takes time to subduct trench location is not directly constrained by
better, for slabs that are much more massive ocean basins many thousands of kilometers isochrons (cf. Fig. 1A vs. Fig. 1B), but given
than plumes. wide, implying a commensurate age for the a number of equally deep (equally old?) can-
The cartoons of Figure  2 illustrate “verti- slab walls, i.e., older than previous estimates, didate slabs, e.g., CR, CR2, SF1, MEZ, and
cal” slab sinking in a lower-mantle sense, i.e., which did not factor in the excess width of the ANG, those closest to the Farallon-Pacific
of thickened slabs. In panel C1, two slab walls, walls. Old age is important to our hypothesis spreading ridge should represent Farallon
ANG and CR, initially grow under two station- because the slabs, if old enough, could not have lithosphere. The Farallon-Pacific ridge ran just
ary, intra-oceanic trenches. By the time of panel subducted beneath the western margin of North east of reconstructed Pacific isochrons, as in
C2, the Angayucham Ocean has closed, and its America—the continent lay too far east during Figure 4A, where the ridge is backstripped to
arcs have accreted to migrating North America, Jurassic times, when it was still part of the super- 110 Ma. The slabs closest to it are CR, CR2,
while the ANG slab wall continues to sink verti- continent Pangea or just breaking away. Hence, and SF1, and hence these should be Farallon
cally in the mantle. The CR (Farallon) arcs have the significance of widened slab walls (regard- slabs—a reasoning confirmed by ongoing Far-
also accreted to the continent, which is now less of the exact widening mechanism) lies in allon subduction into CR.
forcing the Farallon trench westward. Com- the sheer volume of lithosphere contained and SF1 truncates upward at depths of ~1050 km,
bined with vertical sinking, this forced trench the long memory of mantle history recorded. and CR2 truncates upward at depths of between
rollback deposits a laterally extended slab from Their steepness suggests both vertical sinking 1050 and 650  km, as seen in Supplementary
the transition zone downward. Thus, vertical and stationary trench positions over those long Movie M1 (see footnote 1; although these
sinking can generate both vertical and shal- periods of time. numbers may be too shallow because resolu-
lowly dipping slab geometries, depending on Slab wall widening probably occurs through tion tests indicate upward smearing due to the
trench motion. Figure 3, showing a different lithospheric bending and folding when sub- slabs’ residence beneath the uninstrumented
oblique view of CR/Cascadia subduction, and ducting lithosphere approaches the viscosity Pacific; Sigloch, 2011). Hence, subduction into
its comparison to Figures 2A and 4A, permits interface near 670  km depth. This would lay SF1 and CR2 terminated long ago, quite con-

Geological Society of America Bulletin, v. 129, no. 11/12 1497

 Sigloch and Mihalynuk

sistent with the Farallon’s fragmented history as by deposition into the CR slab, which reaches Slabs CR2 and K point to a problem with the
inferred from isochrons. Sager et al. (1988) re- sufficiently deep (>1800  km), is voluminous Andean analogue in that they must represent
constructed a clockwise rotation of the northern enough (widened to 400–700 km), and connects Farallon slabs (being most proximal to the re-
Farallon plate ca. 147 Ma, suggesting a change upward to ongoing Farallon subduction. Suf- constructed Farallon spreading ridge; Fig. 4A).
in trench strike from roughly E-W to its more ficiently deep and voluminous slabs (SF1, L1) Yet, they could never have subducted beneath
recent NW-SE direction. Indeed, CR and CR2 can also account for southern Farallon subduc- the continental margin because even today they
strike NW-SE, but at depths below ~1400  km tion (which may or may not have initiated as are located west of the west coast. So at least
(red shades in Fig. 4A), an E-W–striking con- early as 180+ Ma). Even known isochron com- these old parts of the Farallon plate must have
nection between CR and CR2 is imaged, which plications such as rotations and plate break-offs subducted offshore, raising the question: What
we interpret as Farallon slab deposited be- are matched by slab geometries. else did?
fore this rotation (the implied sinking rate of
1400 km/147 Ma is 9.5 mm/yr, consistent with
the estimate of 10 ±  2  mm/yr by Sigloch and
Mihalynuk, 2013). Following rotation, the co- Figure 4 (on following page). Reconstruction of North American drift and its impingement
existence of CR and CR2 en echelon was pre- on the Archipelago of Jurassic–Cretaceous island arcs, with the resulting assemblage
sumably not stable, and CR2 subduction ceased of accreted arc terranes and continent-spanning suture of the Mezcalera-Angayucham
in the Cretaceous. When SF1 died out, a suitable Oceans. (A) Superposition of subducted slab walls (as in Fig. 2A) with quantitative plate re-
shallower slab, L1, is imaged further east, which construction of Müller et al. (2008) in the lower-mantle reference frame of Steinberger and
would have accommodated subsequent Farallon Torsvik (2008). Paleo-position of North America is in black; seafloor isochrons are in dark
subduction. L1 is a slab that fills the transition blue. Reconstructed time is 110 Ma; only slabs presumably deposited by that time are ren-
zone at ~400–800 km depth, which is too shal- dered (and are currently located below 1100 km depth, based on a sinking rate of 10 mm/a;
low to be rendered in Figure 2A or Figure 4A, Sigloch and Mihalynuk, 2013). Colored text labels and lines are used to group genetically
but it is outlined in dashed green in Figure 4A, related tectonic elements: in orange, the Mezcalera slab wall (MEZ), interpreted Mezcalera
and it can be inspected in Supplementary Movie trench locations, and the Mezcalera Ocean (westward-subducting); in red, the Angayucham
M1 (see footnote 1). Initiation of L1’s trench is slab wall (ANG), Angayucham trench, and ocean (westward-subducting). These inferred
marked as the easternmost green trench barb oceans fill the space between westward-migrating North America and the MEZ and ANG
in Figure 4A; the slab itself is discussed as the slabs, as outlined by the cyan-colored patches. All other slabs and trenches, labeled green,
“Laramide slab” by Sigloch (2011), although are associated with the eastward-subducting Farallon plate. CR, CR2, L1, SF1, K—see sec-
we now think that it spans a wider time range tion 2.4. Paleo-arcs are being extinguished and accreted where North America has started
than the Laramide flat subduction episode. to impinge on the eastward-projecting MEZ slab (dashed orange barbs), followed by sub-
Judging by its depths of 900–1200 km, slab K duction flip to eastward below the newly established Andean-style margin (slab L1, solid
may be the direct “successor” of CR2. Its E-W green barbs). (B) Current versus former positions of accreted terranes. North America’s
strike and location north and west of CR (which position is reconstructed for 170 Ma, 140 Ma, and present day; one intermediate position
represents the northern Farallon plate) make of western margin at 90  Ma is shown by dark green coastline contour. Arc terranes are
it tempting to identify K with the Kula plate shown in their current locations (left half of plot) and in their inferred paleo-positions, i.e.,
(Fig. 3A). Isochrons indicate that the Kula plate behind the intra-oceanic trenches of panel A, and above the slab walls, which are outlined
broke away from the northern Farallon plate and as translucent patches. The same color scheme as in panel A is used to group related tec-
subducted northward between ca.  85  Ma and tonic units across time periods: in orange, Mezcalera slab, ocean, paleo-arcs, and accreted
ca.  55  Ma (Woods and Davies, 1982; Enge- terranes (Insular and Guerrero Superterranes); in red, Angayucham slab, ocean, arcs and
bretson et  al., 1985; Atwater, 1989). While terranes (Alaska and Siberia); in green, Farallon slabs, arcs, and terranes; in purple, Inter-
slab K’s limited depth span of 300  km would montane Superterrane (IMS), which had accreted by 170 Ma. The superterranes in their
be consistent with the Kula plate’s 30 m.y. life present positions are subdivided as described in Table 1: in orange, MEZ-affiliated arc ter-
span and a deposition rate of 10 mm/yr, its ab- ranes PE (Peninsular), WR (Wrangellia), AX (Alexander), WF (Western Jurassic, Western
solute depth exceeds the 550–850 km expected Hayfork, Foothills, and related terranes), SA (Santa Ana), and GU (Guerrero). The purple
for a 10 mm/yr sinking rate. On the other hand, terranes QN (Quesnel), ST (Stikine), CC (Cache Creek), and YTT (Yukon Tanana) are all
this slab is not underlain by a massive wall, and considered part of Jurassic IMS Superterrane. Equivalents in the conterminous U.S. are
an argument could be made that it fell through shown together as BM (Blue Mountains terranes), which are linked to the Native Jurassic
the upper mantle at a much faster velocity. If arc (NJ) and its continuation into Mexico (“Nazas arc,” diagrammatically shown by purple
slab K is not Kula lithosphere, it could instead asterisks). The collapsed basins that mark the Mezcalera-Angayucham suture are shown in
represent northward subduction prior to Kula solid cyan, also annotated by numbers and listed in the legend. These basins represent the
breakoff. During the Cretaceous superchron surface remains of the large paleo-oceans shown as translucent cyan patches, and straddle
(ca.  120–83  Ma), spreading was not recorded the boundary between orange (MEZ) and purple (IMS/NJ) terranes as predicted. Between
magnetically, and the transition from (north- basin relicts 4 and 8, the Insular-Intermontane boundary is extensively overprinted by
ward) Izanagi to (northward) Kula subduction is Coast Cascades orogen. Where adjacent arc terranes were expansively below sea level, su-
poorly known (Woods and Davies, 1982). ture basin strata may have extended well beyond the suture zone, such as northeast of suture
Thus, all Farallon seafloor known to have ex- relict 5, where coeval Bowser Basin (BB) is observed as onlapping Intermontane Superter-
isted can be accounted for by the western half of rane (IMS). Sierra Nevada batholith, located immediately east of basin 10, overprints the
the observed slab assemblage (slabs CR, CR2, basin and the Native Jurassic arc. Green terranes are associated with Farallon subduction:
K, SF1, L1). At least 180 m.y. of northern Far- CG—Chugach; PR—Pacific Rim; SC—Siletz-Crescent; FR—Franciscan; VC—Vizcaino.
allon subduction can be directly accommodated Abbreviations and terrane attributions are explained further in Table 1.

1498 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

A SW
A4
NE 0
km

A2
1000
A1

O
rc

AN Bas
as

G in
NE
Izanagi A1

CR
Ocean

CR
2

L1
A2
A4

MEZ
Mezcale
SW
Fara
llon

Farallon

ra Ocea
Pacific
-Pa

Ocean
Ocean
cifi

SF 500 km
c s

n
1
pr

1000 km
e
ad
in

rid 1500 km
g

ge
Figure 4 2000 km

B
?
PE, ?
WR,
YTT, ?
AX 1 2 ST, Angayucham
QN,
CG
CC Ocean
Angayucham (ANG) 3
4
slab & arc
BB
Mezcalera (MEZ) 3 Al
slab & arc as
5 6 ka
IMS & native Triassic- n
* PR 7
Jurassic arc in south- ar
west USA & Mexico SC 6 cs
ANG/MEZ basins 8
IM S

BM
& sutures WF 9
arcs

Farallon lithosphere / NJ
IM S

subduction complex FR
10
Cascadia
Insular

Suture basins with / without NJ

mantle or accretionary complex Root *
1. Tlikakila *
2. Kahiltna VC
* * *
Guerre

*
3. Gravina-Nutzotin 11

*
SA
4. Kluane schist
5. Anyox GU *
Mezcalera
Ocean
*
6. Tyaughton-Methow
ro arc

7. Bridge River, Hozameen

8. San Juan, Shuksan, et al.
9. Mitchell Inlier
s

10. Hornbrook, Galice,

Mariposa, Don Pedro
11. Arperos: Taray, El Chilar

Geological Society of America Bulletin, v. 129, no. 11/12 1499

 Sigloch and Mihalynuk

2.5. MEZ and ANG Slab Walls Cannot all lower-mantle slabs of Figures  2A and 4A westward subduction, because no slabs are im-
Be Farallon Lithosphere initially grew under separate, but coeval, intra- aged to the east of MEZ and ANG, and because
oceanic trenches (the archipelago), including MEZ/ANG cannot represent Farallon litho-
Strikingly, the matching of Farallon seafloor the green Farallon trenches. sphere, as argued earlier herein.
to subducted slabs leaves the huge MEZ and As North American rode over Mezcalera-
ANG slabs unaccounted for. Associating MEZ 2.6. Archipelago Override— Angayucham trenches and into the archipelago,
and ANG with Farallon lithosphere is implau- Geophysical Predictions it encountered Farallon lithosphere or smaller
sible for other reasons as well: microplates. Subduction regimes for this litho-
(1) Wrong depth—The Farallon plate could Supplementary Movie M1 visualizes the four- sphere were variable and complex but can be
not have subducted into CR, CR2, SF, and into dimensional model predictions for archipelago roughly inferred from slab geometries. In Fig-
the more distal MEZ/ANG slabs at the same override (see footnote 1). As North America ure 2A, southerly cross-section X-X′-X″ slices
time. If MEZ and ANG were Farallon slabs, migrates westward from 200  Ma to 0  Ma in through the MEZ slab at a latitude where it
they would need to represent older Farallon 5 m.y. increments, the slab is rendered (depos- was overridden by the southern U.S./Mexican
subduction, despite residing at the same depths ited) at increasingly shallow depths, at a rate of west coast. Panels B1 and B2 of Figure  2 in-
as CR, CR2, and SF. This implies vastly slower 10.5 mm/yr (see section 2.7 for rate derivation). terpret observed slab geometries along this sec-
sinking or complete stagnation of MEZ and Interpreted trenches are placed vertically above tion: Westward closure of the Mezcalera Ocean
ANG, despite being the more massive slabs. actively growing slabs and evolve with the slab (panel B1) was followed by a forced subduction
(2) Slab longitudes—A coast-hugging Faral- geometries. Figure  5 summarizes this trench flip to eastward Farallon subduction (B2). While
lon trench would have swept across all longi- evolution from 200  Ma to present. The first- the stationary, intra-oceanic Mezcalera trench
tudes between the 170 Ma margin and today’s order contrast is between older trenches hover- had built a vertical slab wall, the migrating,
coastline (as reconstructed in Fig. 4B). Yet, be- ing stationary above slab walls versus younger margin-hugging Farallon trench built a dipping
neath much of this area, no slabs are observed, trenches dispersing in a westward-younging pat- slab L1 (which is too shallow to be rendered in
especially not beneath the cyan-shaded areas of tern. In all cases, the transition coincides with the Figure  2A but can be inspected in Movie M1
today’s Atlantic Ocean. arrival (override) of North America at the slab [see footnote 1] and Fig. 5).
(3) Steep, narrow geometries—A marginal walls, i.e., trench override. Westward subduc- Northerly section Y-Y′-Y″ in Figure 2A cuts
trench would have constantly migrated west- tion into stationary MEZ is replaced by eastward through the two equally deep slab walls ANG and
ward with the continent, but the near-vertical subduction into the westward-migrating Faral- CR indicating a subduction history quite differ-
MEZ/ANG slab walls suggest stationary lon (L1) slab. Stationary CR transitions into the ent from that of section X-X′-X″. Both the ANG
trenches, over sufficiently long times to have de- west-migrating Gorda slab; both are deposited and the Farallon trench were intra-oceanic and
posited the massive volumes observed. via eastward subduction of Farallon lithosphere. stationary (panel C1). After closure of Angayu-
(4) Slab shape—The outlines of the MEZ and Override of the archipelago must have taken cham Ocean and extinction of Angayucham arc,
ANG slabs do not conform to the shape of the roughly 100  m.y. This geophysical predic- there was still no subduction beneath the west
North American margin, against expectations tion follows from the east-west extent of the coast (because no shallow slab equivalent to L1
for slabs deposited at the margin. MEZ, ANG, and CR slab walls in Figure  4 is imaged between ANG and CR; this micro­
(5) Uniform extent in depth—Both the ANG (~2800 km, measured at latitude 65°N), divided plate presumably escaped to the north, with
and MEZ slab walls extend equally deep at by the rate of North American westward drift, margin-parallel motion). The CR (Farallon)
their eastern and western limits (Figs.  2A and which is well constrained by Atlantic spreading slab wall continued to grow until the west coast
4A; cross section in Fig.  4A inset for MEZ; (~25  mm/yr over the past 155  m.y., measured impinged on its trench and started dragging it
Fig. 5 and Movie M1 [see footnote 1]), sug- at latitude 40°N; e.g., Engebretson et al., 1985; westward (panel C2). This initiated the Andean-
gesting simultaneous deposition rather than Müller et  al., 2008). This rough averaging of analogue subduction regime along the Cascadia
westward-younging deposition by a migrating, (2800 × 106 mm)/(25 mm/yr) yields 112 m.y. In margin and has since built a dipping slab that
marginal trench. a more granular accounting that propagated ob- connects to the current trench, which is clearly
(6) Slab curvature—The eastward-protruding servational uncertainties, Sigloch and Mihaly­ observed in panel A, and is cartooned in panel
shape of the MEZ-ANG chevron is inconsistent nuk (2013) predicted onset of archipelago over- C2. Hence, the conversion to Andean-style sub-
with long-lived eastward subduction because ride at 146 ± 24 Ma when the North American duction happened much later along Y-Y′-Y″
this would have created a space problem for the margin reached point A1 in Figure  4A (the than along X-X′-X.″ At even more northerly
incoming Farallon plate; i.e., had the MEZ‑ANG easternmost extent of MEZ slab/arc). Thus, first latitudes, the conversion is not expected to have
structure been a singular Farallon arc, its “cor- collisions would have been in the Late Jurassic. happened at all.
rect” curvature would be westward convex. Tomography and plate reconstruction predict end Such predictions make our hypothesis testable
In summary, if the North American slab walls of archipelago override at 55 ± 7 Ma, when the and falsifiable. Spatio-temporal collision predic-
were generated by continuous eastward sub- margin reached the CR trench. tions can be checked against the geologic record
duction beneath the continental margin since On its continued westward drift past the first (e.g., Dickinson and Lawton, 2001; Dickinson,
180+  Ma, then there is no recognizable geo- MEZ arcs at point A1, the west coast diachro- 2004, 2006, 2008; Sigloch and Mihaly­nuk,
metrical signature of such a relationship. This nously closed the wedge-shaped Mezcalera 2013). If no matching collisions or sutures are
problem disappears if the MEZ and ANG slabs Ocean from north to south, and the Angayu- observed, then slab sinking must have been more
were of intra-oceanic and non-Farallon origin. cham Ocean from south to north, accreting complex than just vertically down, and error
Intra-oceanic trenches could have remained sta- increasingly broader swaths of Mezcalera and bounds can be put on these deviations.
tionary in the seas west of Pangea and deposited Angayu­cham arc terranes to the south and north, The geologic deformation record indicates a
the observed, steep slab walls. Making these respectively (Figs. 4A and 4B). The Mezcalera- completely consistent time span for override:
most easterly slabs intra-oceanic implies that Angayucham Oceans must have closed by onset of southern Canadian Rocky Mountain

1500 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

Figure 5. Volcanic arc and trench

migrations west of Jura-Cretaceous
North America over time and space:
summary of the four-dimensional A
paleo-reconstruction of Supple-
mentary Movie M1. Colored lines
are absolute trench positions over
K
**

AN
time, with respect to the lower
*

G
mantle, as inferred from slab geom-
etries. ­Colors are chosen to match *

CR
the depth-to-color mapping used G
*
*

CR
to v­isualize the 3-D tomography

L1
Depth & age

2
model in Figures 2 and 4, and Movie

EZ
0 km 0 Ma
*

M
M1 (see footnote 1). At depth incre-
*

SN
extents of accreted

*
ments of 100 km, trench geom­etries and native Triassic /

*
were inferred from the geome­ 670 km Jurassic arc rocks

*
tries of “actively growing” slabs, as 1000 km ~100 Ma

explained in the caption of Movie ~140 Ma

ME Z
M1 (see footnote 1). A slab-wall *

SF
sinking rate of 10.5 mm/yr, which 2000 km ~190 Ma
*

1
is estimated from slab geometries
F/C
rather than just assumed, produces Subducted lithosphere Slab Slab deposition style
a mapping of slab depth to trench & polarity outlines & sinking rate
age; see the color bar. This mapping Angayucham 800 km supported, ~10.5 mm/a
is derived from the geometries of Farallon/Cocos 1400 km
unsupported, >10.5 mm/a

the MEZ, ANG, and CR slab walls

(c.f. Section 2.7); such “supported”
Mezcalera
Pacific ? stagnant, <10.5 mm/a

trenches are drawn as solid lines. ?

Barbs on trench lines point in direc-
tion of subduction; barb colors denote the ocean basin origin of subducting lithosphere: Farallon (dark green), Mezcalera (orange), ­Angayucham
(red), or Pacific (maroon). Slabs that sank slower or faster than 10 ± 2 mm/yr are marked by dotted or dashed trench lines, respectively, and their
deviating sinking styles are expected and discussed in the caption of movie M1 (see footnote 1). Arguments for MezAng suturing do not depend
on these younger Farallon slabs. Slab outlines at 1400 km and 800 km depth are given for reference, as are continent locations in a lower-mantle
frame at 170 Ma, 80 Ma, and present day, with accreted belt shaded beige. (Additional times are omitted in order to avoid clutter. The movie
reconstructs 3-D slab deposition, continental drift, and seafloor isochrons in time increments of 5 m.y., from 200 Ma to present.)

deformation at ca.  163–146  Ma (Pană and gin protruded at least 200–300 km farther west 2.7. East-Dipping Slab Geometries Support
van der Pluijm, 2015), and the Nevadan orog- than today (relative to stable North America), Archipelago Hypothesis and Provide Rate
eny in California 163–152  Ma, terminating ­considering the previous addition of the Inter- Estimates for Slab Deposition
around 155 ± 3 Ma in the Sierra Nevada foot- montane microcontinent to the margin in Canada
hills (­Schweickert et al., 1984) or 145–155 Ma (Evenchick et al. 2007, and references therein), The eastward-protruding chevron where
(Harper et al., 1994), although some work- subsequent shortening in the Rocky Mountain the MEZ and ANG slabs meet is probably the
ers posit ongoing deformation for ~30 m.y., to and Skeena fold-and-thrust belts, and similar most discussed geometric feature of the North
ca. 123 Ma (Tobisch et al., 1989; Saleeby et al., minimum cumulative shortening in thrust belts American slabs and has been interpreted to
1989), significantly overlapping the onset of of the United States (e.g., DeCelles, 2004). In support Andean-analogue subduction, includ-
Sevier orogenic events (e.g., DeCelles, 2004). that case, the margin would have impinged on ing our own work prior to Sigloch and Mihaly­
This matches our predictions of deformation on- MEZ arc ~10–15 m.y. earlier, consistent with nuk (2013). Tomography models have typically
set (collisions at MEZ promontory → ­Nevadan the age of the Nevadan orogeny (Schweickert been presented in two-dimensional (2-D) cross
orogeny), followed by more interior defor- et al., 1984; Harper et al., 1994). sections that cut obliquely to the strike of ­either
mation as MEZ-ANG arcs were overridden Also, depending on the width of these pre-­ slab wall, most often east-west sections (an
(→ ­Sevier). The last arc terrane accretion along accreted terranes (shaded belts in Fig. 4B and 5; example is Fig.  1 in Sigloch et  al., 2008, cut
the Cascadia margin at 55–50 Ma (Sigloch and and Movie M1 [see footnote 1]) at the latitude at 42°N). This gives MEZ the appearance of
Mihalynuk, 2013) corresponds to a fundamen- of California, override of MEZ (cross section a thick, eastward-dipping sheet, i.e., a dipping
tal change in the Coast-Cascade orogenic belt X-X′-X″ and panels B1/B2 in Fig.  2) should upper surface paralleled by a dipping lower sur-
(e.g., Crawford et al., 2009). have commenced between 135 Ma and 110 Ma, face. A section parallel to the MEZ wall reveals
The temporally least constrained match consistent with the onset of both the main pulse a different geometry, as shown by the inset of
is for deformation onset: 146 ±  24  Ma from of Sierra Nevadan magmatism (­ Ducea, 2001; Figure  4A, which strikes SW-NE. The upper
geo­physics versus 163–145  Ma from geology. Cecil et al., 2012) and the Franciscan accretion- surface of the MEZ slab plunges from SW to
The likely reason is that the continental mar- ary complex (Dumitru et al., 2010). NE, from depths of 500  km (light-blue level)

Geological Society of America Bulletin, v. 129, no. 11/12 1501

 Sigloch and Mihalynuk

down to 1500  km (red). The bottom surface, active, so that 10 mm/yr is also the rate at which corresponds to the Gulf of Carpentaria–Arafura
however, runs almost flat, at ~1800  km depth the slab wall grew in “height”: ~10  km were Sea, etc. Hence, the southwest Pacific provides
(magenta level). The Andean analogue suggests added every 1  m.y., or 100  m.y. of depo­si­tion rather detailed predictions and actualistic ex-
no reasonable explanation, but this geometry is to build a wall to a vertical extent of 1000 km. amples for the geologic structures expected in
expected if the lower surface records the onset (This rate does not hold for upper-mantle slabs the Cordillera if the Archipelago hypothesis is
of MEZ subduction, synchronous along the en- that are not supported from below by the correct, such as arc accretions, subduction flips,
tire sectioned segment, because a continuous lower mantle.) and formation of interior seaways (e.g., Silver
plate boundary came into being (see Movie M1 Figure 4A is calibrated to 10 mm/yr sinking and Smith, 1983). The analogy is detailed fur-
[see footnote 1]). Subduction into this trench rate: matching the reconstruction at 110  Ma, ther in Table 2, which compares the sequences
segment would have been roughly NW-ward only slabs at and below 1100 km depth are ren- of North American versus Australian archi­
(trench-perpendicular). Any cross-section in NW dered. In the same manner, Movie M1 shows pelago override side-by-side.
direction would perpendicularly inter­ sect the reconstructions and tomography slices in incre- Section 3 focuses on geological arguments
MEZ wall, which would look like the cartoon ments of 5 m.y. and 50 km (see footnote 1). for the existence and diachronous closure of
in Figure  2B1. However, a section through A1 the inferred Angayucham-Mezcalera Ocean. Its
would show earlier subduction termination than 2.8. Archipelago Override—Today’s counterparts are the Indian Ocean and Coral Sea
a section through A2 or A4. This illustrates that Southwest Pacific as a Modern Analogue to either side of Australia. Oceanic lithosphere
the slab’s NE-dipping upper surface in the inset ahead (north) of Australia, which formerly con-
of Figure 4A records the diachro­nous extinction Today’s southwest Pacific Ocean is a remark- nected the two ocean basins, has subducted be-
times of subduction along the Mezcalera trench, ably close analogue to the Mesozoic archi­pelago neath the archipelago’s arc terranes, which have
as North America overrode more and more of suggested by tomography and plate reconstruc- now accreted to formerly “native arc” (northern
it (and subduction was forced outboard, e.g., to tions. Some detailed comparison is warranted, New Guinea) as Australia has started to override
trench L1 in Fig. 4A). At A1, the slab wall was given that an archipelago setting is more com- the archipelago. This situation reflects the North
only built to a depth of 1500 km (red color level) plex and has been given less consideration than American archipelago at ca. 110 Ma (Fig. 4A),
and thus a “wall height” of 300  km (= 1500– the Andean analogue, especially by the geo- when the formerly continuous Mezcalera-­
1800  km). At point A2, the west coast arrived physical community. Angayucham Ocean had been separated into
later, so the slab wall was built to the ­yellow The overarching similarity between the two two basins as North America overrode the NE-
level of ~1000  km depth (or “wall height” of archipelagos is that both accommodate(d) the protruding part of Mezcalera trench and its arc
800 km = 1000–1800 km). How long did it take simultaneous and long-lived subduction of two terranes (the Insular superterrane, consisting of
to deposit those 300 km or 800 km of slab wall? major ocean basins from opposite directions. the Wrangellia, Alexander, and Peninsular ter-
The plate reconstruction gives the arrival Their convergent plate boundaries form outer ranes). The phase of Mezcalera-Angayucham
times of the west coast at points A1, A2, etc. In bounds of the archipelagos. Beneath the south- suturing represents the earlier (Jurassic to Early
Figure  4A, for example, the western margin at west Pacific, the Indo-Australian plate subducts Cretaceous) parts of override, equivalent to
110  Ma has overridden the Mezcalera trench northeastward, and the Pacific plate subducts Australia’s past. In Australia’s tectonic future
(dashed orange barb) up to point A2 or A4 (there westward. Beneath the Mesozoic American presumably lies a messy sequence of terrane
is a geological uncertainty of several hundred archipelago, the proto-Pacific Farallon plate accretions, causing frequent reconfigurations
kilome­ters about the exact shape of the margin, subducted eastward, and the Mezcalera-­ of secondary trenches in the archipelago’s inte-
cf. section 2.6). The slab wall sinking rate can be Angayucham Oceans (the North American rior (black barbs in Fig. 6), which will produce
estimated at any point Ai by dividing the (cur- plate) subducted westward. In both cases, one ­significant slab complexity. Figure 5 clearly
rent) depth of the upper surface of the slab wall of the plates carries(carried) a major continent demonstrates the transitions upon override,
by the time since the margin arrived there, e.g., that obliquely overrides(overrode) the archi- from simple to complex trench geometries.
1050 km/110 Ma at point A2. An arbitrary num- pelago and slowly extinguishes(ed) its trenches: Slab geometries in the upper mantle and transi-
ber of sinking rate estimates can be made along present-day Australia plays the same role as late tion zone under North America, which recorded
the MEZ and ANG slab walls, including much Mesozoic North America. archipelago override in full swing, are less vo-
farther west. Sigloch and Mihaly­ nuk (2013) In a remarkable coincidence(?), the connec- luminous and more fragmented than the mas-
chose five points that could be independently tivity of the major, convergent plate boundaries sive, clear-cut, and vertical slab walls that were
verified and dated by the geologic record, for is almost identical for both archipelagos. This built before the continent collided with the first
example, first arc collision and deformation at becomes evident if a map of today’s southwest Mezcalera-Angayucham terranes. Thus, the
the continental margin for A1. This ensemble of Pacific is mirrored and then rotated 90° coun- archi­pelago hypothesis offers a new framework
points yielded relatively uniform sinking rates of terclockwise (Fig. 6). Comparison of Figure 6 for assessing upper-mantle slab geometries,
10 ± 2 mm/yr, i.e., relative regional variations in with Figure  4A (at ca.  110  Ma) shows that which have been imaged consistently and in
slab wall sinking of only 20%. Thus, slab sinking trench topologies are basically the same, which great detail since the advent of the USArray
rates are not assumed by our methodology but is made explicit by the use of identical colors for experiment (e.g., review by Pavlis et al., 2012;
were estimated from observations. Across the corresponding trenches and by labeling today’s Burdick et al., 2014), and which can for exam-
lower-mantle slab assemblage, whether MEZ, southwest Pacific map with the names of its ple be inspected in Supplementary Movie M1
ANG, or CR, only this one relatively uniform corresponding Mesozoic American elements of (see footnote 1). Interpretation of upper-mantle
sinking rate is consistent with the geophysical Figure 4. The Angayucham Ocean corresponds structure is beyond the scope of this study, as is
observations and the vertical sinking hypothesis. to the Indian Ocean; the ANG arc terranes cor- a detailed discussion of corresponding geologic
Although 10 mm/yr is the (averaged) sinking respond to Sumatra-Java; the Nevadan and events since the Late Cretaceous (ca. 110 Ma),
rate since end of arc activity, the slab walls pre- ­Sevier orogenies correspond to the Papuan fold- or of slab geometries imaged under the south-
sumably sank at the same rate while the arc was and-thrust belt; the Cretaceous Interior Seaway west Pacific.

1502 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

of appropriate age straddle the boundary of two

Asia microcontinents, which are introduced in sec-
Angayucham Pacific tion 3.1: the Intermontane superterrane, pre-

Siberia Basin Ocean accreted to the continent by ca. 170 Ma, and the

Insular-Guerrero superterrane, accreted as the
Mezcalera-Angayucham Oceans sutured. Sec-
tion 3.2 distributes the inventory of North Amer-
Indian
Future Ocean Australia
ican terranes along the geophysically inferred
trenches (slab walls) such that the predicted
Alaska override sequence reproduces the tectonostrati-
graphic relationships of the accreted Cordilleran
terrane assemblage.
Interpretations of the Cordilleran geologic
record have mostly focused on arguing con-
Orcas sistency with the Andean-analogue model. A
Basin relatively late (post–Late Jurassic) suture of the

IMS North Atlantic Insular-Guerrero superterrane to the Intermon-

tane superterrane is inconsistent with this model
WIS

America because there was arc activity on the Insular

superterrane, which could not have been due
CR to a continent-hugging Farallon arc if the In-
sular super­terrane was not sutured to the Inter­
INS montane superterrane (the latter had accreted
prior to 170 Ma and hence represents continen-
Gulf of
Mezcalera Mexico
tal margin for our purposes). Section 3.3 argues
that all supposed geologic evidence against
SRC Basin late Insular-­ Inter­
montane superterrane sutur-
W
Central ing is controvertible because none of said evi-
dence demonstrates stitching or overlap sensu
N S
America
E GUS stricto. To the contrary, sections 3.4, 3.5, and
3.6 demonstrate direct observational support for
South
Farallon America
late suturing, and for the preceding Mezcalera-
Angayucham arc activity built atop the Insular
superterrane (by westward subduction), west of
Figure 6. Today’s southwest Pacific is a close analogue to the archipelago inferred for the the suture. Hence, neither land geology nor geo-
seas west of North America in Early Cretaceous times. A 90° counterclockwise rotation of a physical evidence is consistent with Farallon-
mirrored map transforms the larger Australian region (map inset in top right) to a trench- beneath-continent subduction since 180+  Ma,
continent configuration that is topologically almost identical to that inferred from the slab or with the early (pre–mid-Jurassic) accretion
geometries of Figure 4A for Mesozoic North America. (Standard geographic directions N-E- of the Insular superterrane required by Andean-
S-W transform to W-S-E-N; see compass rose in bottom left.) Present-day tectonic elements analogue models of the Cordillera.
are labeled with the names of their corresponding Cretaceous analogues of Figure 4, so that
Australia becomes “North America,” the Indian Ocean becomes the “Angayucham Basin,” 3.1. Cordilleran Superterranes
etc. Trenches are marked by the same colors as their counterparts in Figure 4. Secondary
trenches without counterparts are colored black. Corresponding geological elements and Table  1 lists the characteristics and mutual
their roles in archipelago override are explained in Table  2. IMS—Intermontane super- relationships of all superterranes and terranes
terrane; INS—Insular superterrane; GUS—Guerrero superterrane; SRC—Shatsky Rise discussed, and for which paleopositions and
conjugate; WIS—Western Interior Seaway. current positions are shown in Figure  4B. In
the Andean-versus-Archipelago controversy,
decisive geologic evidence must be held by
3. GEOLOGICAL EVIDENCE FOR THE be rejected based on direct geological evidence the Insular superterrane in the Canadian and
MESOZOIC ARCHIPELAGO MODEL for the Mezca­lera-Angayucham Oceans and for Alaskan Cordillera. Consisting of the Penin-
its spatio-­temporal closure as predicted by the sular and Alexander-Wrangellia terranes, the
From geophysical observations alone (sec- paleo­trench configuration of Figure 4A. Insular superterrane is a Paleozoic–Mesozoic
tion 2), we can ultimately only make plausible Geologic relicts of the Mezcalera-Angayu- terrane composite constructed from several
arguments about paleogeography: Uncer- cham Oceans are preserved in a tract of at least generations of arcs. Located inboard (east)
tain mantle rheologies mean that the scenario 11 collapsed basins extending from Alaska to of the Insular superterrane, there are the “na-
of much more complicated, heterogeneous Mexico. These had been described previously, tive Triassic–Jurassic arc” (Dickinson, 2008)
slab sinking (under an Andean-style trench) but they had not been recognized as a continu- and the Intermontane superterrane, another
cannot be definitively rejected. This section ar- ous, continent-spanning suture of a wide and Paleozoic–Mesozoic microcontinent consist-
gues that the Andean ana­logue can, however, mature paleo-ocean. As predicted, basin relicts ing of Quesnellia, Stikinia, Yukon-Tanana,

Geological Society of America Bulletin, v. 129, no. 11/12 1503

 Sigloch and Mihalynuk

TABLE 2. ANALOGIES IN ARCHIPELAGO TECTONICS AND COMPARISON OF EARLY CRETACEOUS NORTH AMERICA TO TODAY’S SOUTHWEST PACIFIC
Figure 4 Figure 6
Corresponding tectonic elements
Seas west of Cretaceous North America Southwest Pacific today
North America (NAM) at ca. 120 Ma. Australia (AUS) today.
Green: Farallon trenches. Green: Pacific trenches (Izu-Bonin, Mariana).
Orange: Mezcalera trenches. Orange: Coral/Solomon Seas trenches.
Red: Angayucham trenches. Red: Indian Ocean trenches (Sunda-Java-Banda).
Westward spreading of central Atlantic Ocean since ca. 170 Ma. Northward spreading of Southern Ocean since ca. 95 Ma (Brown et al., 2003).
Mezcalera-Angayucham Ocean, closed by westward subduction under the Neo-Tethys Ocean, closed by north/northwestward subduction under the
archipelago. Subdivided into Angayucham and Mezcalera Oceans, following archipelago. Subdivided into Indian Ocean and Coral/Solomon Seas, following
impingement of North America on the archipelago. impingement of Australia on the archipelago.
Farallon Ocean in the proto–Pacific basin, subducting eastward. Pacific Ocean, subducting westward.
Mezcalera and Angayucham trenches formed a continuous and stationary eastern Sunda-Java–Banda Trench forms a smooth, continuous and stationary western
boundary of the archipelago (as evidenced by continuous, steep slab walls). boundary of the archipelago (as evidenced by continuous trench lines and steep
slabs underneath).
Farallon Ocean trenches formed the ragged western boundary of the archipelago. Pacific Ocean trenches form the ragged eastern boundary of the archipelago.
Eastward subduction into slab fragments CR, CR2, SF1, L1. Westward subduction into the fragmented Izu-Bonin, Mariana, Yap, and Pulau
trenches.
Shatsky Rise conjugate (SRC), the subducted other half of today’s Shatsky Rise, Ontong-Java Plateau, world’s largest oceanic plateau on the Pacific plate.
a huge oceanic plateau on the Farallon plate.
Western Interior Seaway (WIS) Arafura Sea/Gulf of Carpentaria
Prior to archipelago override
The archipelago was bounded by two major oceans subducting in opposite The SW Pacific archipelago is bounded by two major oceans subducting in
directions: the Farallon Ocean versus the Mezcalera-Angayucham Ocean (and opposite directions: The Pacific Ocean versus the Neo-Tethys Ocean (and its
its successor basins). successor basins, including the Indian Ocean).
North America rifted from supercontinent Pangea (African margin) to drift Australia rifted from supercontinent Gondwana (Antarctic margin) to drift northward,
westward, opening the central Atlantic Ocean in its wake. opening the Southern Ocean in its wake.
Ahead of NAM, the Mezcalera-Angayucham Ocean closed, by westward Ahead of Australia, the Neo-Tethys Ocean closed, by northward subduction under
subduction under the archipelago. the archipelago.
On its westward drift, NAM first accreted the microcontinent IMS (Intermontane On its northward drift, AUS first accreted the outer Melanesian arc (sensu Cloos
superterrane) ca. 185 Ma. et al., 2005) of Papua New Guinea (PNG) as ocean crust at the leading edge of
AUS was totally consumed by northward subduction.
Archipelago override
North America’s impingement on the archipelago divided the Mezcalera- Australia’s impingement on the archipelago divided the Neo-Tethys into two basins
Angayucham Ocean into two basins on either side of the continent: Angayucham on either side of the continent: the Indian Ocean and the Coral/Solomon/Tasman
Ocean and Mezcalera Ocean. These two basins closed over the next 100 m.y. Seas. These two basins continue to close as Australia continues to ride into the
as NAM rode into the archipelago. archipelago.
Collision of INS produced the Mezcalera-Angayucham suture and a continent- Melanesian arc collision generated a suture and the continent-verging Papuan
verging fold-and-thrust belt (Nevadan and Sevier orogenies). fold-and-thrust belt.
Riding into the archipelago, NAM accreted the active arcs behind the red ANG Riding into the archipelago, AUS will continue to accrete active arc terranes behind
trenches (future Alaska) and behind the orange MEZ trenches (INS-GUS the red trenches (Sunda-Java-Banda) and orange trenches (New Britain, New
microcontinent). Collisions were diachronous because the MEZ-ANG trench lines Ireland, Solomon Islands). Collisions will be diachronous because the trench
(slabs) were eastward convex. lines are curved, mostly southward convex.
Since NAM did not subduct, its collisions with MEZ-ANG arcs forced subduction flip Since AUS will not subduct, its collisions will force subduction flip, from northward
from westward to eastward. This started outboard of INS and IMS, at latitudes to southward. This has already started in northwest (outboard) PNG with
of the United States, initiating the Franciscan subduction complex ca. 125 Ma. subduction of the Pacific plate.
Override of archipelago interior and subduction of its lithosphere (e.g., the small Override of archipelago interior and subduction of its oceanic lithosphere will be
Orcas basin) was messy, as reflected by complex slab geometries in the upper messy; many small trenches and basins exist already.
mantle.
From the Farallon realm, no continent approached. However, two major oceanic From the Pacific realm, no continent is approaching, but the world’s largest oceanic
Farallon plateaus, the Shatsky Rise conjugate (SRC) and Hess Rise conjugate, plateau (30-km-thick Ontong-Java) is headed for the archipelago.
impinged on the archipelago and ultimately on the NAM margin (after ca. 90 Ma).
Presumable cause of the Laramide orogeny was resistance to subduction of
buoyant plateau.
Cretaceous Western Interior Seaway was formed by crustal thickening of IMS and Epicontinental “interior seaway” Arafura Sea/Gulf of Carpentaria between AUS and
continental terrace, causing loading of the continental margin via fold and thrust Papua New Guinea is formed by crustal thickening of continental terrace via fold
deformation (Beaumont et al., 1993; Evenchick et al., 2007) during accretion of and thrust deformation and accretion, causing loading of the continental margin
INS. (Hamilton, 1979; Edgar, 2003).
Thick clastic blankets were shed from the structurally thickened belt, eastward Thick clastic blankets (molasse) are shed southward from structural highlands
across a broad alluvial plain and coal-forming swamplands and into the Western across the broad alluvial plains and swamps of the Fly-Strickland lowlands and
Interior Seaway (e.g., Pană and van der Pluijm, 2015). into the Arafura Sea (flysch).
Completion of Archipelago override
NAM took ~100 m.y. to override the archipelago, accreting all terranes to its AUS might ultimately override the entire archipelago, accreting all terranes to its
western margin. Finished with the override of the most distal Farallon trench, northern margin. Before fully colliding with eastern Asia, Australia might override
CR, at ca. 55–50 Ma. Final accretion did not result in a flip in polarity, but the most distal Pacific trenches, Yap, Mariana, and Izu-Bonin.
conversion from intra-oceanic to margin-hugging trench.
Note: The table refers to Figures 4A, 4B, and 6, listing corresponding tectonic elements and their respective roles in the two archipelago override sequences.

1504 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

and the Cache Creek terranes. Like today’s MEZ slab, so that the Mezcalera-Angayucham Angayucham suture spatially tracks the oro­
Aleutian, Kuril, and Sumatra-Banda arcs, the Oceans closed between the Intermontane super­ clinal folding of the Alaskan arcs.
inner, Quesnellian part of the Intermontane terrane and Insular superterrane while build- West of the Insular superterrane, override of
super­terrane extended from ocean in the north ing the MEZ slab wall by westward subduc- the CR slab is similarly predicted for 60–50 Ma
onto the continent in the south, i.e., the arc on tion (Figs.  4A and 4B). Placement of Insular (Sigloch and Mihalynuk, 2013). In good agree-
Quesnellia transitioned into the native arc an- super­terrane along the intra-oceanic MEZ slab ment, the last major arc accretion occurred
chored in today’s southwest, cratonic United permits a sensible placement of the remaining ca. 50 Ma with the Pacific Rim terrane (Cowan,
States (e.g., Asmerom et  al., 1990; Lawton Cordilleran superterranes: 2003; Groome et al., 2003; Jakob and Johnston,
and Garza, 2014). Hence, the Intermontane (1) Arc terranes that now make up Alaska 2015). Much of this Farallon/CR arc material
superterrane was bound to the native arc, and must always have been located north of the might currently be underplated (Clowes et  al.,
offshore portions began to accrete to cratonic Insular superterrane. This is satisfied if Alas- 1987), still to be tectonically exhumed as the
North America by ca.  185  Ma (Nixon et  al., kan terranes are attributed to formation above modern Cascades accretionary prism continues
1993; Murphy et  al., 1995; Colpron et  al., the ANG slab—a huge slab that is a plausible to be underthrust and accumulates. Later and
1998). Final collapse of the Inter­ montane generation region for Alaska’s voluminous, further outboard, oceanic plateau rocks of the
super­terrane against the margin (at least Jurassic–­Cretaceous arc rocks (not including Siletz-Crescent terrane accreted ca.  48–40  Ma
loosely) occurred by ca.  173  Ma in the north Arctic Alaska and the northward-translated (McCrory and Wilson, 2013). They must reflect
(Mihalynuk et al., 2004) and ca. 170 Ma at the Alaskan parts of the Insular superterrane). early-stage subduction along the (Andean-style)
latitude of California (Dickinson, 2008). After (2) The Guerrero superterrane of Mexico, Cascades arc that replaced the intra-oceanic CR
that time, well-defined arc magmatism did not considered to be a southern extension of the arc. Hence, the Siletz-Crescent terrane could
affect the ancestral North American margin, Insular superterrane, must always have been be thought of as the top/oldest part of the Cas-
only the more outboard Insular super­terrane. located south of the Insular superterrane. This cades accretionary prism. Magmatic rocks of
Hence, the crucial question becomes: Where is satisfied if the Guerrero superterrane was lo- the northern Cascades are as old as 35 Ma (Phil-
were the Insular superterrane and its trench lo- cated atop the southern MEZ slab, immediately lips et al., 1989), although ancestral Cascade
cated after 170 Ma, relative to the North Ameri- south of the Insular superterrane. arc precursors in California and Nevada date to
can/Intermontane superterrane margin? If they (3) The most recent accretions of (Farallon) ca. 45 Ma (du Bray et al., 2014).
were offshore and separate from the pre-ac- terranes must have been associated with the From arc geology in western Mexico, Dick-
creted Intermontane superterrane, then an ocean westerly CR slab and trench. The Chugach sub- inson and Lawton (2001) inferred Cretaceous
intervened between the Insular superterrane duction complex and Pacific Rim arc terrane can westward subduction of more than just a narrow
and Intermontane superterrane (our Mezcalera- be explained as having grown above the east- backarc basin under the Guerrero superterrane.
Angayu­cham Oceans), and its suture must be ward-subducting CR. Assignment of the Guerrero superterrane to the
sought between those two superterranes. By Items 1 to 3 express the plate-tectonic under­ same topological position as the Insular super-
contrast, if the Insular superterrane was bound standing that relative inboard/outboard posi- terrane, only farther south along the MEZ slab,
to the Intermontane superterrane and hence the tions of major terrane belts should not have is consistent with substantial, late westward
craton since ca. 170 Ma (van der Heyden, 1992; changed over time, despite deformation and subduction because the geometrically inferred
or even since 200+ Ma), then no Mezcalera- variable margin-parallel translation. They also Mezcalera Ocean widens to the south (Fig. 4B)
Angayucham Oceans existed, and subduction express the expectation that some substantial arc and thus must have closed later there than be-
beneath the composite Insular-Intermontane terrane should be matched to each of the huge tween the Intermontane superterrane and Insular
super­terrane margin must have been eastward. slab walls. superterrane to the north.
The latter alternative is a key assertion of the The terrane configuration surrounding the
Andean model. Alternative scenarios, such as Intermontane and Insular superterranes passes 3.3. No Geological Evidence for
ours, are not “Andean” because they imply a some important geological plausibility checks. Intermontane-Insular Superterrane
gap in arc activity along the continental west Much of Alaska consists of Jurassic–Creta- Suturing Prior to the Late Jurassic
coast while Jurassic–Cretaceous westward sub- ceous island-arc terranes and basins (red in
duction beneath the offshore Insular super­ Fig.  4B), which overprint and overlap older The decisive test between archipelago versus
terrane was closing the Mezcalera-Angayucham crustal fragments. The presence and growth of Andean-style subduction history is the timing
basin. Intermediate scenarios have invoked these arcs are causally explained by the pres- of Intermontane-Insular superterrane suturing,
simultaneous subduction beneath a marginal ence of the ANG slab. As North America slowly near the narrow nexus of the Mezcalera-Angayu­
trench and beneath the Insular superterrane, and obliquely overrode the ANG trenches from cham Oceans. It should be post–ca.  155  Ma
either eastward (e.g., Monger et  al., 1982) or SE to NW, ending only ca.  55  Ma according (the reconstruction of Figure  4B predicts it
westward (e.g., Dickinson, 2013), and thus at to slab geometries and plate reconstructions was u­ nder way by 140  Ma), or post–146  Ma
least a small intervening ocean. (Sigloch and Mihalynuk, 2013), it would have ± 24 Ma, when considering geophysical timing
deformed and translated the accreting Alaskan uncertainties and uncertainties about the west-
3.2. Location Constraints on Intermontane arc terranes along its margin. This prediction ward extent of pre-accreted the Intermontane
and Insular Superterranes— is consistent with paleomagnetic evidence superterrane (Sigloch and Mihalynuk, 2013).
Regional Context for compaction of the Alaskan arcs through In the Canadian Cordillera and northwest
variable rotations in oroclinal folds, and with Washington State, the Intermontane-Insular
It follows that the archipelago model must northward transport by hundreds to a few thou- superterrane contact is located in the strongly
assign the Intermontane superterrane to the sand kilometers between 85  Ma and 55  Ma metamorphosed and extensively intruded Coast-
North American margin but the Insular super- (Hillhouse and Coe, 1994; Johnston, 2001). Cascades orogen. The true extent of this Coast-
terrane and its Jurassic–Cretaceous arc to the Hence, the Angayucham part of the Mezcalera- Cascades orogen suture is continental in scale

Geological Society of America Bulletin, v. 129, no. 11/12 1505

 Sigloch and Mihalynuk

because the two superterranes once flanked much include arc, ocean crust, and mantle substrate tain ultramafic lenses interpreted as ­slivers of
of the U.S. and Canadian margins (Fig. 4B). Dur- (purple number labels in Fig.  4B). They have mantle (Mihalynuk et al., 1994).
ing and after accretion, parts of the Intermontane the correct distribution, age, and composition Near the southern end of the Alaska pan-
and Insular superterranes were translated, mainly for surface remnants of the closed Mezcalera- handle, surrounded by plutons in the heart of
northward according to paleomagnetic evidence Angayucham Oceans—the suture we seek, al- the Coast belt, is the Anyox pendant. It is ex-
(Enkin, 2006; Kent and Irving, 2010; Hillhouse most in plain sight. tensively faulted and in part composed of a vol-
and Coe, 1994), with the Insular superterrane Starting in the north, in south-central Alaska, canogenic succession of early Middle Jurassic
now flanking British Columbia and Alaska, and much of the border between the Insular and and perhaps Early Jurassic age, including tholei-
only discontinuous relicts of the Intermontane Inter­montane superterranes is marked by the itic pillow basalt and serpentinized ultramafite
super­terrane still present in the United States, Late Jurassic to Early Cretaceous Kahiltna as- (Evenchick and McNicoll, 2002). The clastic
e.g., the Blue Mountains and Klamaths (Dickin- semblage in a region that has long been recog- strata overlying the Anyox pendant are corre-
son, 1976, 1979; Coney et al., 1980; Silberling nized as a tectonic suture (Jones et  al., 1982; lated with those of the Late Jurassic and Early
et al., 1992; and many others). Csejtey et al., 1982) and has thus been named Cretaceous Bowser Lake Group (Evenchick
Our proposed, relatively young Intermon- the “Alaska Range suture zone” (Ridgway and McNicoll, 2002), which contain open-­
tane-Insular superterrane suture in the Coast- et al., 2002). Detrital zircon analysis shows that water faunas (Poulton et al., 1994) and were af-
Cascades orogen is widely considered to be the Kahiltna Basin received detritus from the In- fected by the same east-directed, mid- through
negated by geological evidence. Monger (2014) sular superterrane, and increasingly in the early Late Cretaceous deformation that formed the
summarized the arguments against Late Juras- Late Cretaceous, from the Intermontane super- Skeena fold-and-thrust belt (Evenchick et  al.,
sic–Early Cretaceous suturing as: terrane (Hampton et  al., 2010). Kahiltna has 2007), a consequence of Intermontane-Insular
(1) There is scant evidence of a suture of the been regarded as a syncollisional basin (Ridg- superterrane suturing.
correct age in the Coast-Cascades orogen, where way et al., 2002; Hampton et al., 2007) formed In southwest British Columbia and north-
evidence is limited to the Bridge River terrane. by suturing that proceeded from east to west ern Washington State, basinal strata akin to the
(2) There is evidence of plutons stitching the (Kalbas et  al., 2007), or from south to north Gravina-Nutzotin extend through the eastern
Intermontane-Insular superterrane boundary by in Jurassic coordinates. This is consistent with Coast Cascades orogen to the coeval Tyaughton-­
the Early Jurassic. diachronous metamorphism of Kahiltna basin Methow Basin (McClelland et  al., 1992;
(3) There is evidence of overlap between the strata ranging from Late Cretaceous to early Hampton et al., 2010). Early Cretaceous strata
Intermontane superterrane and Insular superter- Paleocene, and evidence of older and ~14  km near the top of the basin succession (the Relay
rane by the Early Jurassic. deeper metamorphism in the east (Davidson Mountain Group) are recognized to be the old-
(4) The Intermontane superterrane and Insu- and McPhillips, 2007). est probable link between the southeastern and
lar superterrane can be correlated by ca. 180 Ma The Kahiltna Basin displays along-strike southwestern Coast belts and thus between the
on the basis of temporal and faunal affiliations. equivalence with the Nutzotin and Gravina Ba- Intermontane and Insular superterranes (dated
In subsections 3.3.1–3.3.4, we argue that all sins, which extend from eastern Alaska through ca.  110  Ma by Garver, 1992; but revised to
four points are controvertible, because none western Yukon (known there as the Dezadeash ca. 130 Ma by Umhoefer et al., 2002).
of the evidence constitutes a stitch or overlap Formation) and along the Alaska panhandle, At the base of the Methow Basin, the ­Cayoosh
sensu stricto. well into the Coast belt (e.g., McClelland et al., assemblage consists of Early Jurassic to Early
1992; Manuszak et  al., 2007; Ridgway et  al., Cretaceous, mainly deep-water strata that con-
3.3.1. Evidence for the Mezcalera- 2002). Gravina and its metamorphosed equiva- formably overlie oceanic crustal units of the
Angayucham Suture lents separate magmatic belts that, prior to Bridge River Ocean complex (Journeay and
Far from being limited to the small Bridge 110 Ma, were restricted to either the Intermon- Northcote, 1992). Cayoosh strata have been
River terrane or even to Canada, the predicted tane superterrane or the Insular superterrane interpreted as the last vestiges of the Mississip-
Mezcalera-Angayucham suture runs the length but were overprinted by 100–50  Ma plutons pian to Jurassic (Cordey and Schiarizza, 1993)
of the North American Cordillera. Although that span the boundary (fig. 5 in Gehrels et al., Bridge River Ocean within the southern Coast
the large-scale connections may not have been 2009). Intermontane superterrane and Gravina belt (Mahoney and Journeay, 1993; Journeay
made, the predicted trail of Late Jurassic to Early Belt strata are interleaved by thrusts (Rubin and Mahoney, 1994). We regard this long-lived
Cretaceous collapsed basins has been observed et al., 1990; McClelland et al., 1992) and syn- Bridge River Ocean as part of the Mezcalera
by previous workers. It extends from Alaska kinematic, mid-Cretaceous plutons (Crawford Ocean. Other workers seeking to accommodate
and Canada (Hampton et  al., 2010) through et  al., 1987; Himmelberg et  al., 2004) that such evidence of old oceanic crust between the
the southwest United States (McClelland et al., may have acted to facilitate and enhance thrust- Insular superterrane and Intermontane superter-
1992; Anderson and Mahoney, 2006) to Mexico ing (Hollister and Crawford, 1986). Possible rane have proposed sinistral strike-slip entrap-
(Dickinson and Lawton, 2001; Anderson et al., equiva­lents of these strata known as the Kluane ment of the ocean relict (Monger et al., 1994;
2005). The basins include the Kahiltna and Nut- schist in southwest Yukon contain detrital zir- Gehrels et  al., 2009). While sinistral faults
zotin Basins in Alaska; Dezadeash Formation in cons as young as ca. 95 Ma, with metamorphic undoubtedly exist, invoking regional strike-
Yukon; Gravina, Tyaughton, and Methow Ba- overgrowths of ca. 82 Ma (Israel et al., 2010), slip duplication of the Bridge River–Cayoosh
sins in British Columbia and Washington State; and are structurally interleaved with slices of basin is an incidental complication because
the Mariposa and Galice Formations in Califor- Triassic ocean lithosphere (trondhjemite ages: this Mezca­lera Ocean relict is situated exactly
nia; and relicts of the Arperos Basin in Mexico, 206.3 ± 2.0 Ma to 206.8 ± 3.2 Ma; Mónica Esca­ where it should be—exposed within the Insular-
along with the Taray “Formation” and El Chilar yola, 2015, personal commun.). In the northern Intermontane superterrane suture.
subduction complexes. The bulk of the basin Coast Mountains of northwest British Colum- Farther south, Late Jurassic to Early Cre-
fills consists of turbidite deposits (those basin bia, metasedimentary rocks that are undated but taceous amalgamation of terranes may be re-
relicts are colored cyan in Fig.  4B), but some possibly correlative with the Gravina belt con- corded by quartz-rich overlap strata. Late Juras-

1506 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

sic granitic boulders contained in these strata A reconstruction of the North American margin ages fall within the acknowledged range of arc
provide a maximum overlap age that “links” at 80 Ma is plotted as a green line in Figure 4B, magmatism in both the Insular superterrane and
Wrangellia to Bridge River–Methow terranes which crosses the southern limit of the Insular su- Intermontane superterrane (ca.  210–170  Ma;
(Friedman and Armstrong, 1995; Riddell, perterrane arc and subduction zone, thus predict- see earlier herein), so until facies of Moffatt
1991). However, strata as young as Cenoma- ing final ocean closure, possibly including blue- volcanics can be traced from the Insular super-
nian (ca.  95  Ma) contain radiolarian and fora- schist emplacement, on the southernmost Insular terrane to the Intermontane superterrane, they
minifer fossils indicating open-ocean conditions superterrane around this time. do not strictly constitute an overlap. However,
(Haggart et al., 2011) prior to the onset of Late confident tracking of the volcanic units from the
Cretaceous thrusting (Journeay and Friedman, 3.3.2. No Evidence of Plutons Stitching Intermontane superterrane to the Insular super-
1993). Pervasive faulting within this part of the the Insular-Intermontane Superterrane terrane may not be possible because contacts
Coast-Cascades orogen means that overlap can- Boundary by the Early Jurassic of Moffatt volcanics in the Insular superterrane
not be unequivocally proven, a problem that is Until plutons across most of the Insular-Inter- are described as faults, or unconformities modi-
exacerbated in older, more deformed, and more montane superterrane boundary have been sub- fied by faults (Gehrels, 2001), which contradicts
extensively intruded rocks. jected to robust isotopic studies, arguments for and/or calls into question their “overlap” status
Detrital zircons from the mid-Cretaceous or against pluton stitching will lack credence. (although at a few localities, fault disruption of
Methow Basin strata record no definitive evi- Utilization of cooling ages and multigrain zir- the unconformity appears to be minimal; Brew
dence for provenance linkage to the Precam- con U-Pb dating, or even dating of single zircon and Karl, 1988).
brian North American interior (Haggart et  al., crystals containing xenocrystic cores or mantled One of the best opportunities to demonstrate
2011, Surpless et  al., 2014), which is unex- by overgrowths, obscures the geological story. overlap of the Insular-Intermontane super­terrane
pected if the Methow Basin formed as a margin- To illustrate, multigrain U-Pb zircon age data boundary is the westward extension of the
proximal, Andean-style forearc. By contrast, the were used to argue for an Andean margin in an Inter­montane superterrane–derived, distinctive,
lack of continental zircons is more probable if influential paper by van der Heyden (1992), but chert pebble conglomerates of the Late Juras-
a trench east of the Methow Basin, marked by none of the older ages of stitching plutons re- sic to Early Cretaceous Bowser Basin (“BB” in
the Cayoosh subduction complex, separated and ported by van der Heyden (1989, 1992) could Fig.  4B). Into the Coast belt, these strata have
isolated the basin from North America.  In our be confirmed by subsequent workers. All ages been traced as far west as the Anyox pendant
interpretation, Methow was the forearc basin, had to be revised downward when found to have (Evenchick and McNicoll, 2002). However, in
and Cayoosh was the accretionary prism above been compound ages compromised by zircon the adjacent Insular superterrane, west of the
the westward-subducting MEZ slab. inheritance (Butler et  al., 2006; Gehrels et  al. Coast belt, the nearest coeval conglomeratic
Near the latitude of the Canada-USA border 2009). Pluton ages now stand at 100–120 Ma, strata (Gravina belt) are dominated by igneous
(49°N), Jurassic to Early Cretaceous strata of the down from previously 230  Ma, 120–140  Ma, clasts derived from the underlying Insular super­
Methow trough unconformably overlie ocean- and 100–120  Ma, and consistent with Early ter­rane (Brew and Karl, 1988). Such observa-
ridge–type pillow basalt and gabbro, which Cretaceous suturing, as predicted by tomogra- tions are consistent with the observations of
are in fault contact with belts of serpentinized phy and plate reconstructions (see “Magmatic Monger et al. (1983), who stated that all clastic
­mantle (Ray, 1986). This marks the suture be- Response to Archipelago Override” section rocks of the Gravina-Nutzotin belt appear to
tween the deformed leading edge of the Inter- later herein). have westerly sources. Conclusions drawn by
montane super­terrane (Mount Lytton complex these early authors still apply: The simplest solu-
included in Quesnellia) and outboard gneissic 3.3.3. No Conclusive Evidence for Overlap of tion, and best fit to available data, is that the In-
rocks of the Coast-Cascades orogen. the Insular Superterrane and Intermontane sular superterrane accreted to the Intermontane
South of 49°N, the southern ends of the Superterrane by the Early Jurassic super­terrane late, trapping the Jurassic–Creta-
1500-km-long Coast-Cascades orogen and of Examples of Insular-Intermontane super- ceous Gravina-Nutzotin b­ asin in between.
adja­cent Wrangellia (Insular superterrane) are terrane overlap basins older than the Middle
curiously enveloped by Late Jurassic to Creta- Jurassic do not exist, nor can the Late Jurassic 3.3.4. No Correlation via Temporal Affiliation
ceous rocks. In the west, this includes ophiolite and Cretaceous basins discussed earlier be dem- of Arc Ages or Faunal Assemblages
and blueschist in the thrust nappes of the San Juan onstrated to form incontrovertible “overlap.” The Intermontane superterrane and Insu-
Islands, e.g., rocks overprinted by 124 ± 0.7 Ma Rather, they are collapsed remnant basins, in the lar superterrane should be expected to display
blueschist metamorphism (mid–Early Creta- sense of Ingersoll et al. (2003). All have under- an overlap in magmatic ages. Intermontane
ceous; Brown et  al., 2005). The nappes are in- gone postdepositional structural overprinting, as superterrane Stikinian arc subduction was ac-
terpreted to have been emplaced over Wrangel- a consequence of collision that outlasted sedi- tive until termination ca.  173  Ma in the north
lia terrane from the south or possibly southeast mentation. Overlap could only be demonstrated (Mihalynuk et al., 2004), and perhaps slightly
(Brown, 2012) and may record Insular superter- if strata depositionally rested on both the Insular later in the south (Cordey et  al., 1987). West-
rane–Intermontane superterrane collision. Ter- superterrane and Intermontane superterrane and ward subduction into the MEZ slab (beneath the
minal San Juan deformation is bracketed by the could be traced from one to the other. Insular superterrane) should have overlapped
age of the youngest strata involved in thrusting Perhaps the most-cited example of Early Juras- the last ~20–30 m.y. of Intermontane superter-
(112–115  Ma, late Early Cretaceous) and the sic overlap is the Moffatt volcanics. Isotopic age rane subduction, according to slab depths and
oldest postthrusting strata (ca. 84 Ma, Late Creta- dates reported from allegedly correlative portions volumes (Sigloch and Mihalynuk, 2013). Thus,
ceous). The uppermost thrust nappe contains de- on the Insular superterrane (177 ± 4 Ma, “ca. 175 we do not agree with workers who assert that
trital zircons as young as 87 Ma (Brown, 2012). Ma”; Gehrels, 2001, p. 1589) and Intermontane the similarity of the Insular superterrane and
Late suturing at low latitudes is predicted by our superterrane (173 ± 8 Ma and 170 ± 5 Ma) are Intermontane superterrane arc ages, or even
kinematic model, which shows a wedge-like coeval only near the limits of their large error synchronous initiation of arc segments, implies
widening of the Mezcalera Ocean to the south. envelopes (Gehrels, 2001). In addition, these formation along the same margin. Although

Geological Society of America Bulletin, v. 129, no. 11/12 1507

 Sigloch and Mihalynuk

instances exist where this is demonstrably the oceanic crust, and mantle with local lenses of 3.5. Insular Superterrane as the Substrate
case, one need not look beyond the modern-day blueschist. An overlying conglomerate that for the Mezcalera Arc
western Pacific to see counterexamples. In the contains Na-amphibole grains and clasts from
case of Early Jurassic North America, cessation Bridge River blueschist provides the first deposi- Magmatic and tectonic lulls recognized in
of eastward subduction beneath western Pangea tional ties across the Intermontane-Insular super­ land geology are compatible with separation of
(Intermontane superterrane and native arcs), and ter­rane boundary in the mid- to Early Cretaceous the Insular superterrane from the Intermontane
initiation of westward-pulling subduction (into (Garver, 1992; Umhoefer et al., 2002). superterrane before the Late Jurassic (and hence
the Insular superterrane and Alaskan arcs by the In Alaska, Kahiltna strata are at least partly location of the Insular superterrane above the
MEZ and ANG slabs) were probably even geo- deposited unconformably atop the Silurian to first outboard slab to the west). Proponents of
dynamically linked, because both would have Triassic Mystic subterrane, which includes fos- the Andean-margin hypothesis might argue that
facilitated the breakup of Pangea and central siliferous limestone, conglomerate, phosphatic since ca.  207  Ma in the Late Triassic (Amato
Atlantic spreading. shale, well-preserved pillow basalt, mélange, et  al., 2007b), the Insular superterrane mag-
Jurassic faunal affiliations shared by the Inter- and turbiditic sandstone (Bradley et  al., 2007; matic arc in Alaska, the Talkeetna arc, had been
montane superterrane and Insular superterrane Kalbas et  al., 2007). Paleozoic fossils of the coupled with a subduction complex outboard
have been used to argue that the Insular super- Mystic subterrane are more closely allied with of the Insular superterrane, the Chugach ter-
terrane and Intermontane superterrane (Stikine) Siberian than North American faunas (Blodgett rane, which includes blueschists dated between
arcs were built in close proximity by ca.  200– et al., 2002), and detrital zircons are not easily ca. 204 and 185 Ma (Roeske et al., 1989), thus
180  Ma (Smith, 2006). Faunal affiliations are interpretable, displaying mixed North American proving eastward subduction beneath both ter-
primarily controlled by water temperatures and and Siberian provenance (Bradley et al., 2007). ranes throughout this period. This is one of the
ocean currents. We will not speculate on currents Such evidence helps to discriminate between most persistent myths supporting an Andean
in the Mezcalera Ocean, but we point out that archipelago paleogeography and slightly modi- margin in Alaska. In fact, high-pressure–low-
the Intermontane superterrane and Insular super- fied Andean-analogue models (for the United temperature metamorphic mineral assemblages
terrane would have been located at similar lati- States—McClelland et  al., 1992; Saleeby, occur in two modes: within the McHugh com-
tudes on either side of Mezcalera Ocean, so that 1983; Harper et  al., 1994; for Canada—van plex, the innermost (oldest) part of the Chugach
surface-water temperatures should have been der ­Heyden, 1992), which explain the string of terrane; and as slivers within the Border Ranges,
similar. Despite its substantial width, Mezca­lera ­Jurassic basins as transtensional retro-arc basins a crustal-scale fault system that separates the
Ocean did not span two thirds of Earth’s equator, to eastward Farallon subduction beneath the Chugach complex from the Insular superterrane.
as the proto-Pacific (Panthalassa) did, so that Tri- Insu­lar superterrane arc, formed subsequent to Blueschists are important markers of fossil sub-
assic to Cretaceous endemism across it should a hard collision of the Insular superterrane with duction zones, but in the Chugach terrane, they
have been proportionally less pronounced, with the Intermontane superterrane by Early Jurassic predate the assembly of their supposedly associ-
eastern and western faunal affiliations largely times. We are unaware of any other case where ated accretionary complex by tens of millions of
shared. Hence, ocean fossil faunas are unlikely a continental-scale orogenic scar reopened along years and are, therefore, clearly reworked from
to discriminate between the Andean analogue its exact suture location. Still, this scenario might a much earlier high-pressure–high-temperature
and our archipelago model. account for the persistence of these basins at the event. Even the inner McHugh complex, pre-
boundary of the Intermontane superterrane and sumably the oldest part of the Chugach accre-
3.4. Exotic Basin Substrates Indicate Insular superterrane, and possibly even for their tionary complex (with individual structural pan-
Large, Long-Lived Mezcalera- deep-sea nature, but the exotic suture substrates els of sandstone having maximum depositional
Angayucham Oceans and open-ocean faunas of the Bowser Basin ages as old as 169 ± 2 Ma; Amato et al., 2013),
(Poulton et  al., 1994) deposited on the Inter- is constrained by detrital zircons to have been
There is additional, direct evidence for the montane superterrane, and of the Methow Basin assembled after 146 ± 5 Ma (Amato and Pavlis,
Mezcalera and Angayucham Oceans and a late (Kleinspehn, 1985) deposited between the Inter- 2010; Clift et al., 2012), and hence >50–60 m.y.
suture. The oceans’ large dimensions and lon- montane superterrane and Insular superterrane after the oldest blueschist metamorphism. Simi-
gevity are indicated not just by the voluminous are difficult to reconcile with the retro-arc model. larly incompatible ages are seen 200 km south
MEZ and ANG slabs and by direct size esti- If relicts of Late Jurassic to Early Cretaceous of Anchorage, where maximum depositional
mates from plate reconstructions (Fig. 4B), but basinal strata were deposited in transtensional ages of 160  Ma at Iceberg Lake (Day et  al.,
also by exotic substrates that underlie some of or backarc basins, their substrates should domi- 2011) contrast with a 185  Ma blueschist age
the Mezcalera-Angayucham suture basins. Ex- nantly be extended portions of the adjacent Inter- (Sisson and Onstott, 1986).
otic substrates are well demonstrated outside of montane superterrane and Insular superterrane, Initiation of subduction recorded in the
the Coast belt, where metamorphic overprinting not seafloor that formed up to 170 m.y. earlier Chugach complex followed termination of Tal-
and destruction of fossils are not extensive. On containing exotic fauna. In principle, exotic ter- keetna arc volcanism around 167  Ma (Amato
either end of the Coast-Cascades orogen, exotic rane fragments might have been trapped during et al., 2007b; although plutonism may have ex-
substrates sit stratigraphically and/or structur- an earlier episode of suturing and re-exposed by tended to ca. 153 Ma according to Rioux et al.,
ally beneath remnant ocean basin strata: Bridge formation of backarc or transtensional basins. 2007). Our model suggests that the Talkeetna
River terrane beneath Methow Basin strata (in However, in that case, evidence of earlier sutur- arc was not paired with the Chugach complex,
southern British Columbia), and Mystic sub- ing should be demonstrated by age-appropriate but rather grew above the west-subducting MEZ
terrane beneath Kahiltna strata (in the Alaska regional deformation, extending the length of slab that drew the Intermontane superterrane
Range suture zone). the continent, like the Jurassic–Cretaceous ba- toward the Insular superterrane (similarly pro-
The Bridge River accretionary complex con- sins. Such evidence seems to be lacking until posed by Dickinson, 2004). Diachronous Insu-
tains an old oceanic record spanning ~170 m.y. Mezcalera-Angayucham suturing in Late Juras- lar-Intermontane superterrane collision, starting
(Cordey and Schiarizza, 1993) of pelagic chert, sic to Cretaceous times. in the Late Jurassic, terminated Talkeetna arc

1508 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

growth and forced a subduction jump outboard 20  m.y., starting with terminal collapse of the 2001). Since this magmatism directly followed
of the arc, and a polarity flip to (north-) eastward Intermontane superterrane (174–170  Ma) and the end of Cache Creek subduction (Mihaly-
subduction of the Farallon plate. This explains ending with reestablishment of arc magmatism nuk et al., 2004), it is tentatively related to slab
the subsequent onset of Chugach complex ac- after continental override of the Insular super- break-off.
cretion, recording Andean-style subduction of terrane ca. 155 Ma (we attribute the initial pulse Despite a paucity of magmatic rocks between
the Farallon realm, but only since the latest Ju- of Nevadan deformation [Harper et  al., 1994] the Middle and latest Jurassic, continuous vol-
rassic, not since the Late Triassic. to first impingement of the Insular superter- canic ash air fall is recorded in Bowser Basin
rane into North America and its pre-accreted strata by U-Pb detrital zircon ages and by bio-
3.6. Magmatic Response to Intermontane superterrane). Compared to this chronology of coeval ammonites (Evenchick
Archipelago Override tectonic lull between arc collision events, the et  al., 2010). However, recognizable ash beds
magmatic lull on the Intermontane superterrane are extremely rare in the basin, suggesting a
Magmatism is a faithful recorder of plate appears to have been somewhat shorter, which distant source, such as the approaching Insular
inter­actions. Arc magmatism records subduc- can be ascribed to three factors: old, mislead- super­ter­rane volcanic arc. It was not until the
tion. Termination of arc magmatism and at- ing geochronology (as described earlier herein); end of the Early Cretaceous that intrusions be-
tendant deformation record collision with a predictable slab break-off magmatism on the gan to abundantly cut the Bowser Group strata
buoyant crustal welt. Unfortunately, normal Intermontane superterrane; and Intermontane (e.g., Wheeler and McFeely, 1991; Mihalynuk
arc magmatism can be challenging to distin- superterrane magmatism related to deforma- and Friedman, 2006; Evenchick et  al., 2007).
guish from subduction-termination magmatism tion rather than subduction. During the mag- This plutonism is consistent with eastward
caused by asthenospheric inflow after slab break matic lull in the Intermontane superterrane, Farallon subduction that initiated after Insular-
(e.g., van de Zedde and Wortel, 2001) or from arc magmatism in the Insular superterrane was Inter­montane superterrane collision and the
orogenic magmatism caused by crustal thicken- ongoing, as expected from westward Mezcalera forced flip of subduction polarity (Fig.  4A:
ing and collapse (Dewey, 1988; Li et al., 2016) subduction. dashed orange trench turns into solid green
and/or superimposed convective delamination Demonstrating the existence of a magmatic trench near the MEZ-ANG arc chevron). In
(Houseman and Molnar, 1997; Molnar and lull in the Intermontane superterrane is easiest summary, bona fide arc magmatism in the Inter-
Houseman, 2013). We focus on spatio-temporal within a 300  km stretch where the magmatic montane superterrane on the continental margin
predictions for arc activity as determined from slate was wiped clean by the Bowser Basin ended before 172 Ma and resumed only around
slab geometries, which guide the scrutiny of the (labeled “BB” in Fig.  4B; e.g., Wheeler and 130  Ma. However, during this interval, distal
existing magmatic record and its episodicity. McFeely, 1991; Evenchick et al., 2010). The ash sourced in the active, offshore Insular super-
The lack of a slab to the east of the MEZ- Intermontane superterrane basement of the terrane arc contributed zircons to the sediment
ANG slab walls, beneath the Atlantic Ocean, Bowser Basin includes intrusions as young budget of the Bowser Basin.
predicts an arc hiatus on the North American as 185 Ma, overlain by Aalenian strata (172– Local records of intercollisional magmatism
west coast while it was overriding those lon- 176  Ma) intercalated with 175–172  Ma shal- in the Intermontane superterrane exist. Perhaps
gitudes. However, as described, we also accept low submarine volcanic rocks and their feed- the most comprehensively studied example is
174–170  Ma as the timing for the collapse of ers and volcaniclastic strata (which were the composite Endako batholith, located ~70–
the Intermontane superterrane and its native arc considered the youngest Intermontane super- 200 km south of the Bowser Basin. Its magma-
equivalents against the paleo-Canadian and U.S. terrane magmatism at this latitude of 56.5°N tism apparently ranges from ca. 219 Ma (Late
margins, by eastward subduction. In principle, by Cutts et  al., 2015). These rocks form the Triassic) to ca.  145  Ma (youngest Cretaceous;
the Mezcalera Ocean might have closed by si- substrate for the diachronous sediments of the Villeneuve et al., 2001). The isotopic and geo-
multaneous eastward and westward subduction, Bowser Lake Group (Evenchick et al., 2010): chemical character of the youngest Endako
as envisioned by Dickinson (2013, his fig. 7). deltaic strata, prograding southwestward (from phases is consistent with wholesale remelting of
This would have created arc magmatism along Bajo­cian to Oxfordian, ca. 172–156 Ma) over juvenile arc crust, not mantle contributions from
both the Intermontane superterrane and Insu- basinal turbidites, dominated by chert pebble arc magmatism (Whalen et al., 2001). Interme-
lar superterrane margins right up to the time conglomerates derived from the uplifted Cache diate-aged Endako phases are plagued with dat-
of ocean closure, and it would render the dem- Creek terrane (Intermontane superterrane). ing uncertainties due to selection of multigrain
onstration of a suture and of two separate arcs Post–172 Ma magmatic arc rocks are absent in zircon fractions and the possibility that some of
more challenging. However, this scenario is not the Bowser Basin (Intermontane superterrane), these grains contained inherited cores.
supported by slab geometries, because no addi- while they are abundant in the coeval Insular In the northwest United States, the Idaho
tional “native-arc slab” has been imaged in the superterrane. batholith preserves a geological history much
mantle beneath the region formerly occupied Plutonic rocks intrusive into other parts of like that of the Endako batholith. Oldest phases
by the west coast of Triassic–Jurassic North the Intermontane superterrane are rare during show mantle contributions, but younger phases
America (today’s central Atlantic) at the same the intercollisional magmatic lull in the latest show increasing evidence of melting of subja-
midmantle depths that farther west are filled Jurassic. Following collapse of the Intermon- cent crust (Gaschnig et  al., 2011), indicating
by the MEZ slab. Hence, westward subduc- tane superterrane arc around 173 Ma, a suite of they were not subduction related.
tion should have persisted after termination of ca. 172–170 Ma intrusions cuts the axis of the In the southwest United States, an enormous
eastward subduction under the “native arc.” (A Intermontane superterrane (Cache Creek ter- body of isotopic age data constrains the loca-
deeper, older “native-arc slab” may be present rane). North of the Bowser Basin, their cool- tion and timing of three Mesozoic arc belts,
as expected; see discussion in section 5.5.) ing ages smear to ca. 165 Ma (Fourth of July as summarized by Snow and Scherer (2006),
Can we demonstrate this hiatus of the na- batholith, 59.5°N; Ash, 1994). South of  the Dickinson (2008), Schweickert (2015), and
tive arc, which is negated by Andean-analogue basin at Spike Peak (~55°N), their youngest Saleeby and Dunne (2015). In California, the
models? Duration of the hiatus should be ~15– phase is 166.5 ±  1.8  Ma (MacIntyre et  al., oldest and youngest arc-like belts overlap spa-

Geological Society of America Bulletin, v. 129, no. 11/12 1509

 Sigloch and Mihalynuk

tially in an inboard (eastern) location: A mid- <144 Ma—Snow et al., 2010), and resumption Schweickert (1986); Ingersoll (2008); Dickin-
Triassic to Early Jurassic belt (ca. 240–170 Ma) of eastward subduction, as evidenced by the son (2004, 2008); Johnston (2001, 2008); and
is intruded and overprinted by an arc that ignited Early to Late Cretaceous magmatic inflation of Hildebrand (2009, 2012, 2015). Challenges
ca. 153 Ma with emplacement of post-Nevadan the Sierra Nevada batholith (Cecil et al., 2012; were presented by Dickinson and Lawton
intrusions (such as the Guadalupe pluton; Ernst Ducea, 2001). Figure 5 illustrates the setting of (2001) for Mexico and Box and Patton (1989)
et al., 2009), which peaked around 90 Ma with a the youngest Sierra Nevada arc (“SN”) asso- for Alaska. In some scenarios, westward sub-
spectacular flux of magmatism in the Sierra Ne- ciated with Andean-style Farallon subduction duction may have related to the earlier collapse
vada batholith. Both arcs must reflect eastward into slab L1. The geometry shows how earlier of the Intermontane superterrane against North
subduction beneath the southwest United States: at these latitudes, the MEZ trench and subduc- America, but others related to closure of varia-
The older “native arc” developed across the cra- tion complex would have collapsed against tions of the Mezcalera-Angayucham Ocean ba-
ton, and the younger Sierra Nevada–Peninsular a continental margin hosting the even older sins, and some models have incorporated both.
Ranges arc (a Farallon arc associated with slab native arc.
L1 in Fig. 5) sutured the new crustal addition of Like the Bowser Basin of Canada, the West-
4.1. Archipelago Models
the southern Insular superterrane to the craton ern Interior Basin of the United States holds a
(e.g., Kistler and Peterman, 1978). record of volcanic eruptions near the leading
A third arc of intermediate magmatic ages edge of North America, both as ash beds and Both eastward and westward consumption of
(ca. 193–155 Ma; Dickinson, 2008) is pre- detrital zircons (e.g., Christiansen et al., 1994; oceanic crust has been invoked in models ex-
served in accreted, more westerly terranes and Laskowski et  al., 2013). These reflect cessa- plaining delivery of island archipelago terranes
is separated from the two eastern arcs by relicts tion of vigorous native arc magmatism near to the western margin of North America. Mon-
of blueschist and broken formation that accu- the end Triassic (by 200 Ma) and approach of ger et  al. (1982) posited that the two meta-
mulated outboard of the Intermontane super- the Mezcalera arc (ca.  170–146  Ma, by west- morphic belts extending along the Canadian
terrane/native arc in the so-called compound ward subduction). Magmatism ebbed follow- Cordillera arose from mid-Jurassic and Cre-
suture belt (Dickinson, 2008). Arc rocks west ing Mezcalera collision and then resumed with taceous collisions of two microcontinental ar-
of the suture belt comprise much of the Foot- vigor as new, eastward Farallon subduction was chipelagos: the Intermontane superterrane and
hills and San Pedro terranes (as defined by established outboard of the sutured Mezcalera Insular superterrane, both transported to North
­Schweickert, 2015) and also include the Smart- arc (slab L1 in Figs.  2 and 4). As outlined in America via eastward consumption of oceanic
ville block, Foothills arc, and Slate Creek–Lake Sigloch and Mihalynuk (2013), but not dis- crust (Monger et al., 1972; Monger and Price,
Combie arc belts (of Dickinson, 2008). In the cussed herein, collision of the Hess-Shatsky 1979; Fig. 7B).
Klamaths, the arc includes the Western Juras- conjugate plateau (Livaccari et al., 1981), start- Moores (1970, 1998) hypothesized an intra-
sic, Western Hayfork, and Rattlesnake Creek ing ca. 93 Ma, extinguished the Sierra Nevada oceanic archipelago at the origin of Cordilleran
belts of Dickinson (2008). In detail, the com- batholith, and collision of the Cascadia Root mountain building, which included eastward
pound suture belt is resolved into constituents, arc (starting ca. 55 Ma) extinguished the Coast- (Farallon) subduction as required by the Pacific
like the Calaveras mélange belt of the western Cascades orogen; in both cases, terminal peaks magnetic record, but also westward subduc-
Sierra Nevada range (Dickinson, 2008), and in magmatic flux may have been enhanced by tion beneath an archipelago, ahead of North
to its immediate west, the Don Pedro terrane slab breaking. In Figure 5, the approximate lo- America’s westward migration (Fig.  7C). Far-
(Schweickert, 2015). cation and east-west extent of the Hess-Shatsky ther south, at the latitude of the Sierra Nevada
In our interpretation, the western arc formed plateau during collision are indicated by the batholith in the western United States, workers
above the westward-subducting Mezcalera wide unfilled arrow. have recognized that arc and oceanic crustal
Ocean (in agreement with Dickinson, 2008; remnants and their ages of deformation (“Ne-
Schweickert, 2015). The Calaveras mélange vadan orogeny”) are most consistent with Late
4. EARLIER CHALLENGES
belt and Don Pedro terrane that separate the arc Jurassic collision of an intra-oceanic Jurassic
TO THE ANDEAN-STYLE
from the more easterly, but partly coeval native archipelago that formed above a west-dipping
SUBDUCTION MODEL
arc are thus hypothesized to hold two subduc- subduction zone (Schweickert and Cowan,
tion complexes of overlapping ages: recording 1975; Ingersoll and Schweickert, 1986; Inger-
the Mezcalera basin’s initially double-sided clo- This section puts our findings in the con- soll, 2012; Fig. 7C). Transport to the continen-
sure from the Late Triassic to shutdown of the text of the four different categories of models tal margin was by double-sided consumption of
native arc (ca. 170 Ma, Calaveras mélange belt), found in the geologic literature for Mesozoic ocean crust beneath the archipelago and beneath
finalized by one-sided, westward closure in the Cordilleran evolution: the Andean analogue of the margin of North America.
Late Jurassic (up to ca. 155 Ma, Don Pedro ter- eastward subduction (Fig. 7A); a variant featur- For the seas offshore Jurassic–Cretaceous
rane). This interpretation seems to satisfy both ing staggered eastward subduction (Fig.  7B); Mexico, Dickinson and Lawton (2001) pro-
geological and slab geometrical constraints and archipelago models of westward subduction posed a westward subduction scenario similar
makes the Calaveras mélange–Don Pedro ter- beneath the archipelago (Fig. 7C); and models to ours (Fig. 7C), abandoning an earlier model
rane units part of our continent-spanning Mezca­ of offshore “ribbon continents” that contained of the type in Figure  7A, whereby deposition
lera-Angayucham suture. allochthonous pericratonic terranes (and even of the Mezcalera formation had been attributed
Collision of the Mezcalera arc (Insular super­ craton) of acknowledged North American ori- to westward closure of a narrow backarc basin
terrane) in the Late Jurassic forced subduction gin (Fig. 7D). Challengers of 180+ m.y. of east- (Dickinson, 1981; Goldhammer and Johnson,
outboard of the western arc, as evidenced by ward, Andean-style subduction have included 1999), which was presumed to have opened
the accretionary phase of Franciscan com- Moores (1970, 1998); Schweickert and Cowan due to eastward Farallon subduction beneath
plex growth beginning in the Early Creta- (1975); Chamberlain and Lambert (1985); an Andean-type margin. Dickinson and Lawton
ceous (ca.  123 Ma—Dumitru et  al., 2010; or Lambert and Chamberlain (1988); Ingersoll and (2001) concluded instead that this “Mezcalera”

1510 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

Figure  7. Competing models

for the Jurassic–Cretaceous
A B basin must have been a sizeable, deep ocean—
in good agreement with evidence for the size,
development of the Cor­d i­ timing, and subduction direction of the geo-
llera (A–D), and the oceanic physically inferred Mezcalera Ocean. Like us,
sutures they predict (E). Mod- and supporters of the archipelago model for the
els disagree about collision western United States, they inferred simultane-
times of North America with ous, double-sided subduction in Jurassic times,
Insular (INS) and Guerrero westward under the Guerrero super­terrane (the
(GUS) super­terranes (medium southern extension of the Insular superterrane),
and light gray), and about and eastward under the native arc (called “Nazas
the subduction regimes that C D arc” in Mexico), followed by exclusively west-
closed the surrounding oceans. ward subduction (Fig. 7C). When the Mezcalera
Inter­montane microcontinent IMS basin had been fully consumed, a flip to east-
(IMS) is shown in dark gray. ward Farallon subduction beneath the outboard
(A) Andean analogue model: edge of the Guerrero superterrane ensued (the
long-lived eastward subduction INS modern-day configuration, aside from more
beneath a prefused and pre- recent Gulf of California spreading). Given
accreted Insular-Intermontane the consistency of this Mexican scenario with
superterrane package (e.g., GUS our interpreted paleography further north, we
Burchfiel et al., 1992). Variants: ­adopted their name “Mezcalera Ocean” for the
mid-­J urassic collapse of an entire basin.
ocean basin within Intermon- E
tane superterrane (gray barbs;
Sutures 4.2. Ribbon Continent Models
e.g., van der Heyden, 1992),
or backarc/transtensional Dickinson & Lawton
basin formation along the 2001 (C)
Considering only slab geometries without
Hildebrand 2012-15 (D)
Intermontane-Insular super­ Johnston 2008 (D) additional geological constraints, the joint
terrane boundary (stylized Monger 1982 (B) placement of the Insular superterrane and Inter-
ridge; e.g., McClelland et  al., Schweickert 2015 & montane superterrane behind the Mezcalera arc
1992). (B) Modified Andean-­ many others (C) (i.e., fused since the Early Jurassic or longer)
margin model: collapse of Terrane Belts would be reasonable per se. Including pieces
Insular super­terrane onto Crushed J-K basins of pericratonic terranes farther east would es-
Inter­montane superterrane via Insular sentially equate to the ribbon continent sce-
staggered eastward subduction Intermontane nario of Johnston (2001, 2008) and Hildebrand
Deformed foreland
(Monger et  al., 1972; Monger (2009, 2012), and if limited to north of latitude
west of
and Price, 1979). (C)  Archi­ Deformation front ~48°N, to “Cordilleria” of Chamberlain and
p e l a g o m o d e l s : We s t w a r d Lambert (1985). However, these scenarios re-
subduction of North America quire the Mezcalera-Angayucham suture to run
(with pre-accreted Intermon- east (inboard) of the Intermontane superterrane
tane superterrane) beneath (Fig.  7D). This is negated by geological evi-
Insular-Guerrero superterrane dence of the older, Triassic–Jurassic native arc
(Moores, 1970, 1998; Dickin- rooted in the craton in the southwestern United
son, 2004, 2008; Dickinson and Lawton, 2001; Schweickert and Cowan, 1975; Ingersoll States, which probably connected to Quesnellia
and Schweickert, 1986; Ingersoll, 2008; Sigloch and Mihalynuk, 2013; this study). Vari- on the Intermontane superterrane. Given this tie
ants: Some models consider synchronous or slightly older eastward subduction in addition to the craton, the native arc must have formed
to westward subduction. Not all models consider the Guerrero superterrane. (D) “Ribbon above eastward-subducting ocean lithosphere.
continent” models, including Intermontane superterrane and pericratonic terranes with an Like most workers in the Cordillera, we con-
offshore microcontinent that accreted by westward subduction (Chamberlain and Lambert, sider the evidence for this arc very solid. Plutons
1985; Lambert and Chamberlain, 1988; Johnston, 2001, 2008). Variant: The ribbon conti- along the trail of the native arc cut deformed
nent also included Proterozoic cratonic domains (Hildebrand, 2009, 2012, 2015). Ribbon continent-fringing strata and Proterozoic craton
continents are the only model category that does not have the Intermontane superterrane alike. In addition, Late Triassic detritus from the
pre-accreted by ca.  170  Ma. (E) Ocean suture locations predicted by authors advocating native arc extends atop the craton as far as the
models in B to D. Solid black—Mezcalera-Angayucham suture predicted here (archipelago, Colorado Plateau, e.g., the Chinle Formation
part C), which largely coincides with Monger (1982) in Canada (dashed white, part B) and (Stewart et  al., 1986; Asmerom et  al., 1990;
with Schweickert (2015) and others in California (dashed white, part C), and partly coin- Riggs et al., 2013), and Jurassic eolianites ex-
cides with Dickinson and Lawton (2001) in Mexico (dash-dotted white; part C). Ribbon tend discontinuously from the craton to the
continent models by Johnston (black sawtooth; part D) or Hildebrand (dotted black; part D) native arc in southern California–Arizona and
predict more easterly, inboard sutures within the Intermontane superterrane and/or the northern Sonora (Bilodeau and Keith, 1986;
(peri-)cratonic foreland. Hildebrand suture is pictured as offset by Texas lineament (thin Dickinson and Gehrels, 2009, 2010; Dickinson
dotted). J-K—Late Jurassic–Cretaceous. et al., 2010; Lawton and Garza, 2014).

Geological Society of America Bulletin, v. 129, no. 11/12 1511

 Sigloch and Mihalynuk

Additional issues with ribbon continent sce- displacement is indicated for Early to Late Cre- 5.2. Slab Widening—Observational
narios are that they do not explain the presence taceous Alaskan rocks, which were deposited Robustness and Formation Mechanism
of the Late Jurassic–Cretaceous suture basins along the southeastern ANG and accreted early,
that stretch from Alaska to Mexico (Fig.  4B), whereas weaker and variable, north or south dis- The observed widening of American lower-
and, in the case of Hildebrand (2012), sutur- placements are indicated for latest Cretaceous mantle slab walls to 4–7 times the thickness of
ing times postulated for the Late Cretaceous to Paleocene rocks (Hillhouse and Coe, 1994), subducted oceanic lithosphere plays a crucial
(“Laramide orogeny,” Lewis thrust) are too late which were deposited farther northwest and ac- part in our argument for old and intra-oceanic
to be explained by subduction into the MEZ- creted late. Such timing reflects diachronous, subduction. Several lines of reasoning indicate
ANG slabs. Plate reconstructions confidently southeast-to-northwest closure of the Angayu- that this observation is robust.
superimpose North America with the imaged cham Ocean, in two stages. First, between Late (1) If the slabs were much narrower than
slabs by Late Jurassic–Early Cretaceous times Jurassic to earliest Cretaceous times, the North 400–700 km, then the dominant artifact afflict-
(Figs.  4A and 4B), implying earlier collision American margin ploughed northward and ing ray-based tomography methods would be
with MEZ and ANG arc terranes. Evidence of the obliquely into the trench marked by the ANG wave-front healing. It tends to smear out imaged
earliest phase of this collision is the well-docu­ slab, trapping Angayucham arc terranes and features but acts even more strongly to render
mented Late Jurassic shortening cratonward of their substrates between North America and Si- them invisible (Nolet and Dahlen, 2000). Yet,
the Mezcalera-Angayucham suture (“Nevadan beria. Later, between latest Cretaceous and mid- the slab walls have always been among the most
orogeny”; Harper et al., 1994; Schweickert and Eocene times, North America moved relatively robust features imaged in the mantle, and they
Cowan, 1975; Ingersoll and Schweickert, 1986). southwestward (Enkin, 2006), resulting in dex- are truly massive.
Deformation persisted through Cretaceous times tral strike-slip faulting that accentuated Alaskan (2) Waveform-based or finite-frequency
(“Sevier orogeny”; Armstrong, 1968; Monger oroclines, rotating them 44° ± 11° counterclock- tomog­raphy methods such as we employed here
and Price, 1979; DeCelles, 2004; DeCelles and wise (Coe et al., 1989). Together, these events (Sigloch, 2011) largely compensate for wave-
Coogan, 2006; Evenchick et al., 2007; Pană and produced the “terrane wreck”–style of oroclinal front healing by modeling finite-wavelength ef-
van der Pluijm, 2015; and many others). compaction described by Johnston (2001). fects. Yet, the width of the lower-mantle slabs
Models that focus exclusively on westward has not decreased significantly compared to
subduction beneath a ribbon continent, such 5. DISCUSSION ray-based studies (including comparison tests of
as the “Cordilleria” of Chamberlain and Lam- 5.1. Oblique Collision as the Default our own). Again, this means the slabs must be
bert (1985), “Saybia” of Johnston (2008), and Regime—Not Captured by Andean nor sufficiently wide that waveform healing is not a
“­Rubia” of Hildebrand (2009), are also incom- Ribbon Continent Models severe problem.
plete in that they do not explain how the ribbon (3) Even with ray-based methods, it has al-
continents have related to, and transitioned into, The 3-D slab geometries that do not strike ways been easier to resolve the lower-mantle
eastward Farallon subduction since 180+  Ma, p­ arallel to the reconstructed North American mar- slab walls than, for example, the Juan de Fuca
which is required by the magnetic seafloor rec­ gin indicate that oblique collision was the default plate, just a few hundred kilometers below its
ord on the Pacific plate (e.g., Engebretson et al., state along the Mesozoic west coast. These geom- trench. This should not be the case if the slab
1985; Atwater, 1989; Sager et al., 1988). etries imply formation of slab windows, alternat- had the same thickness of ~100 km in both loca-
Our archipelago model differs from ribbon ing stress regimes, persistent indenters, and mar- tions, because wave-front healing and smearing
continent models in several important aspects, gin-parallel escape, consistent with findings from act more detrimentally on structures far from
most fundamentally, in its constituent parts, synoptic paleomagnetic studies (Enkin, 2006; sources and receivers (i.e., in the lower mantle,
since it includes no western Laurentian peri- Kent and Irving, 2010). At any time between not near a well-instrumented surface region).
cratonic terranes in the archipelago. Still, the ca. 155 Ma and ca. 50 Ma, parts of the continent Slabs in the upper and lower mantle have
term “Cordilleria,” introduced by Chamberlain were colliding with limbs of the MEZ-ANG arc often been imaged by different studies using
and Lambert (1985), is a particularly fitting chevron, while others were not yet colliding, different methods, which might have put this
and self-explanatory name for the intra-oceanic were overriding seafloor inside the archipelago, comparability in question. With the advent of
paleo­geog­raphy that became the western North or were already abutting the Farallon plate. USArray however, individual tomographic stud-
American Cordillera. Hence, “Cordilleria” Although Andean and ribbon continent mod- ies have imaged through both upper and lower
might be readopted as a proper name for “the els are commonly seen as opposite end mem- mantle (e.g., review by Pavlis et  al., 2012;
archi­pelago,” a name inclusive enough to survive bers of Cordilleran tectonic interpretations, they Burdick et  al., 2014). They resolve a sharply
further refinements of the archipelago concept. share the 2-D, special-case character of margin- defined, thin slab near the trench as expected,
With regard to Alaskan terranes, our archi- perpendicular subduction. Both model types and pronounced thickening from the transition
pelago model generally agrees with the “terrane ignore or deemphasize gradual, margin-parallel zone down (e.g., fig. 1 in Sigloch et al., 2008,
wreck” of Johnston (2001, 2008). We associ- changes that arise from unstable plate-boundary transect at 42°N). This striking contrast in slab
ate the ANG terranes with the railcars of his configurations and terrane translations, which character—thin and ragged near the trench
“wreck,” and the ANG slab wall with the abso- are inevitable consequences of plate interactions versus massive and continuous in the lower­
lute spatial location of his ribbon continent. The on a spherical planet. Hence, simple, purely mantle—is also evident in Figure 2A and espe-
slab runs from beneath Nova Scotia to Yukon margin-perpendicular models of subduction cially in Figure 3. A more subtle effect of arti­
and thus grew arcs at more southerly latitudes should not be expected to capture the essence ficial slab widening was described by Bezada
than occupied by (central) Alaska today—in of Cordilleran evolution, at least not on a whole- et  al. (2013), due to unaccounted wave path
good agreement with paleomagnetic con- orogen scale, over 10–100  m.y. To re-create a bending into the slab when using a spherically
straints on Alaska, which generally show north- paleotectonic history that has any semblance of symmetric reference model. In principle, this
ward displacement relative to cratonic North accuracy, there is no way around assimilating effect should affect our images, but in a much-
America.  More specifically, strong northward observations and modeling in 3-D (plus time). attenuated manner because our tomography is

1512 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

dominated by longer wave periods of 30–10 s 1994; Grand et al., 1997; van der Hilst et al., cause total amounts should be the same in both
and uses less approximate sensitivity kernels 1997; van der Lee and Nolet, 1997; Bunge and cases (our model subtracts from the Farallon
than those of Bezada et  al. (2013). Also the Grand, 2000; Ren et  al., 2007; Sigloch et  al., Ocean what it adds to the Mezcalera-Angayu-
North American slab walls are 2–3 times wider 2008; Liu et al., 2008). In part, this is explained cham Oceans; cf. Fig. 1A vs. Fig. 1B).
than their Mediterranean upper-mantle slab and by the relatively late discovery that the surviving To our knowledge, no convection simulation
so are relatively less affected. Farallon plate fragment, the Juan de Fuca plate, has produced the full, 3-D assemblage of North
In summary, lower-mantle slabs are clearly does not connect downward into the MEZ and American slabs discovered since the advent
several times wider than slabs in the upper ANG slabs, as had been assumed, but instead of USArray, and especially not the slabs that
300  km, and our basic argument of massive into the much more westerly Cascadia Root are most problematic to the Andean-analogue
volumes does not depend on whether they are slab (CR; Sigloch et al., 2008), which had been model: the equally deep extent of the ANG slab
4 or 7 times wider. Instead, the existence, width, only vaguely visible in earlier tomographic im- wall at its eastern and western ends, or of the
and age of the Mezcalera-Angayucham Oceans ages (e.g., Bijwaard et  al., 1998; Montelli MEZ wall at its northern and southern ends, nor
are argued from the distances of reconstructed et  al., 2006). Until then, the MEZ and ANG the presence of the deep, westerly CR slab. We
North America to the slab walls, from the tim- slabs had to be presumed to be Farallon slabs expect that generating this full assemblage will
ing of the suture, and from the ocean floor sedi- to match the undisputed seafloor magnetization be nearly impossible with a margin-hugging
ments found in the suture basins; sinking rates for 180+  m.y. of Farallon spreading and thus trench configuration and the known constraints
are inferred from upward truncations of the slab subduction (Engebretson et al., 1985; Atwater, on Farallon spreading, but it would be relatively
walls. While “unfolding” and “obducting” the 1989; Seton et al., 2012). straightforward with the trench evolution we
slabs walls back to the surface might test the Geodynamic simulations of subduction his- propose because many convection simulations
reconstruction of the Mezcalera-Angayucham tory under North America found the observed model do show near-vertical sinking. For ex-
Oceans (e.g., Wu et  al., 2016), we currently slab geometries challenging to reproduce unless ample, Steinberger et  al. (2012) found that in
consider the uncertainties rather large, not just unusual rheological regimes were invoked. Gen- standard rheologies, lateral slab displacement
in the exact slab width, but also the tightness of eration of the pronounced, eastward-pointing from the surface to the core-mantle boundary
its folds, and the likely existence of a spreading chevron of the MEZ-ANG slabs from a rela- did not exceed a few hundred kilometers, i.e.,
ridge in the Mezcalera-Angayucham Oceans tively straight margin-hugging trench proved within the observational uncertainties of verti-
that generated additional lithosphere. especially problematic. Bunge and Grand cal sinking as hypothesized here. Bunge and
Our deductions remain valid regardless of (2000) could produce the chevron shape only Grand’s (1994) model did not produce suf-
how exactly the slabs widened, but recent geo- by inserting the “Farallon” trench along the ficient lateral slab displacement to deposit the
dynamic modeling studies confirm that folding Rocky Mountain thrust front during Cretaceous MEZ/ANG slabs under a marginal Farallon
was the probable widening mechanism. Folding times, an extremely easterly position that they trench. The modeling of Lithgow-Bertelloni
of viscous thin sheets (lithosphere) at the vis- acknowledged as geologically impossible. In- and Richards (1998) enforced vertical sinking
cosity interface between the upper and lower stead, they hypothesized that from its actual beneath trenches globally, including a mar-
­mantle, as shown diagrammatically in our car- coastal trench, Farallon lithosphere had reached ginal Farallon trench, and did not produce the
toons in Figure 2, has been observed in fluid dy- its easterly resting place (MEZ/ANG slab) by observed MEZ-ANG slab chevron (their fig. 6
namic tank experiments (Guillou-Frottier et al., extremely shallow (“Laramide”) subduction, and plate 3).
1995; Bellahsen et  al., 2005; Schellart, 2008) even though their simulations did not produce In a simulation-free approach, van der Meer
and numerical convection modeling (Ribe this lateral translation. Simulations by Liu et al. et al. (2010) assumed vertical sinking globally,
et al., 2007; Stegman et al., 2010; Ribe, 2010; (2008) did achieve shallow Farallon slab trans- including beneath a margin-hugging Farallon
­Gibert et  al., 2012; Čížková and Bina, 2013; port over large lateral distances (~1000 km) by trench. Depositing the steep slab walls then re-
Garel et  al., 2014). The simulations show that implementing a >3000-km-long low-viscosity quired invoking a longitudinal, time-dependent
folding occurs in certain subspaces of a priori channel (“stress guide”) beneath stable North rotation of the global lithospheric shell relative
plausible parameter spaces for slab and ambient America. This modeling complexity was driven to the lower mantle. This net lithospheric rota-
mantle rheologies, which means that observa- by the presumed requirement of a margin- tion would have acted to effectively cancel out
tions of ubiquitous slab walls (hence ubiquitous hugging Farallon trench rather than a priori ex- Atlantic opening during Cretaceous times, in or-
slab folding) and of quantifiable sinking veloci- pectations about anomalous mantle rheologies, der to hold the coastal Farallon trench stationary
ties put strong constraints on mantle rheology. and the extensive low-viscosity layer it implied above the MEZ-ANG slabs, though again with-
­Gibert et al. (2012) suggested that the oscilla- under an old continent is not evident in surface- out fitting the chevron shape of MEZ-ANG slabs
tory motion of slab folding in the mantle tran- wave tomography (e.g., review by Schaeffer (their fig. 2c). This longitudinal shift was argued
sition zone might transmit upward into trench and Lebedev, 2015). from the fit of three “anchor slabs,” but for the
oscillations, possibly expressed in periodically Liu (2014) contrasted slab deposition in the relevant times, only the Farallon plate provides
alternating compressive and extensive stress re- Andean versus archipelago scenarios and ar- a real constraint: The east-west–striking Tethys
gimes along the South American Andes arc. gued for the former. It is, however, unclear how slab yields only weak constraints on longitude,
his 2-D simulations (depth plus a lateral dimen- and the Mongol-Okhotsk slab is too old.
5.3. Andean-Margin Analogue sion) could have implemented the essence of The net lithospheric rotation hypothesis can
in Geophysics the archipelago’s oblique and shifting trenches be tested and rejected by our investigation of
of Figure 5 (no details of trench evolution were MEZ-ANG arc collisions. We show that the
All slab interpretations prior to Sigloch and given). An indication that the essence of our Mezcalera trench was intra-oceanic, which ne-
Mihalynuk (2013) followed the primacy of the model was not captured is that twice as much gates the need to westward-translate the conti­
Andean-analogue model, positing only Farallon lithosphere was subducted than for the Andean nental margin during Cretaceous times. Our
subduction beneath the continent (e.g., Grand, model, a major kinematic inconsistency, be- accretion events indicate no net lithospheric

Geological Society of America Bulletin, v. 129, no. 11/12 1513

 Sigloch and Mihalynuk

rotation with respect to the lower mantle, other number of slab interpretations that have incor- expected in the lower third of the mantle under
than that contained in the hotspot/hybrid refer- porated extensive geological evidence have today’s central Atlantic (at depths correspond-
ence frame we use (O’Neill et al., 2005; Stein- yielded slow sinking rates consistent with ours, ing to the lowest 200 km of the MEZ slab wall
berger and Torsvik, 2008). implying ~10  mm/yr for lower-mantle slabs and deeper). Beyond the reach of the regional
Taken together, the prior work can be read as ­under Asia (Van der Voo et al., 1999a, 1999b) tomography of Sigloch (2011), such ultradeep
having falsified the hypothesis of vertical sinking and 12 ± 3 mm/yr for lower-mantle slabs glob- imaging poses the challenge of adequately mod-
and uniform rheologies given a margin-hugging ally (van der Meer et al., 2010). The latter study eling seismic wave propagation in proximity
trench, but having left open the possibility of ver- scrutinized the geologic record for volcanic arc to Earth’s core, although tomographic imaging
tical sinking for the archipelago alternative. activity (similar to our approach), and although methods are advancing rapidly. We predict that
To our knowledge, the only other slab inter- they assumed vertical sinking rather than treat- this slab will soon be imaged directly, which will
pretations to discuss similarly old arc accretions ing it as a null hypothesis to be tested, we gen- further quantify the spatial and temporal separa-
as ours were van der Meer et al. (2012), who as- erally agree with their results on sinking rates tion of eastward subduction under the Intermon-
sociated Wrangellia (Insular superterrane) with because our tests do support vertical sinking. tane superterrane versus westward subduction
a slab beneath the central Pacific, and Shephard Our sinking estimate was inferred only for under the Insular superterrane. In fact, suitable
et  al. (2013), who acknowledged that the pre- the special-case geometry of massive, steep slab candidates under the central Atlantic Ocean
dicted position of Wrangellia corresponds with slab walls in the lower mantle. We claim no lowermost mantle have started to appear, e.g., in
MEZ, but who supported a conventional inter- valid­ity for other slab types. In fact, Movie M1 the global model of van der Meer et al. (2010,
pretation of MEZ as a post–100  Ma Farallon (see footnote 1) implies clearly slower sink- their supplement) or the compilation of core-­
slab. Both models imply that that Insular super­ ing for Farallon slabs L1 and G sprawling in diffracted waveform measurements for tomogra-
terrane accreted after eastward transport on the the transition zone—the type of “stagnant phy of Hosseini and Sigloch (2015). Despite the
Farallon plate (a modified Andean scenario; slab” first discovered under the archipelagos current lack of confident, direct observations of
Fig.  7B), although consistency with isochron of today’s west and southwest Pacific (­Fukao this paleo-seafloor, our argument about the longi­
data was not argued. For some relatively recent et  al., 2009)—and faster sinking for some tudinal separation of the Intermontane super­
accretions of oceanic plateaus, possible signa- younger Farallon fragments of limited verti- terrane from the Insular superterrane remains
tures left on slabs have been discussed. Accre- cal extent, including the presumed Kula slab K solid because reconstructions of the absolute lo-
tions of the Hess and Shatsky Rise conjugates of section 2.4. There are plausible reasons for cation of Pangea, and thus of the Triassic–Juras-
between 90  Ma and 70  Ma were invoked as their different sinking styles, as discussed in sic native arc on its western margin, are already
explanations for the chevron shape of the “Far- the caption of Movie M1 (see footnote 1). constrained by a range of marine magnetic, land
allon” (MEZ-ANG) slabs (Liu et al., 2010), The present study does not depend on deci- magnetic, and hotspot observations, including
and for the tomographically visible thicken- phering these young Farallon slabs in detail, the backstripping of Atlantic Ocean opening.
ing of the Wyoming craton, via underplating as long as the cumulative volume of Faral- Similarly, the Insular superterrane microcon-
of the Shatsky conjugate (Humphreys et  al., lon fragments is sufficient to account for the tinent has a volcanic arc history that reaches
2015). Accretion of the Siletzia oceanic plateau entire Farallon isochron record, which is the much further back than the accretion event re-
ca.  50  Ma has been associated with a stalled case (section 2.4). The complexity of Farallon constructed here, but again slabs below 2000 km
Farallon slab under the Challis magmatic arc fragments and trench evolution (Fig. 5) is con- depth, which will probably account for this his-
of the northwest United States (Schmandt and siderable but not more so than in today’s south- tory, are only vaguely visible beneath North
Humphreys, 2011). None of these relatively re- west Pacific (Fig. 6). The MEZ/ANG slabs have America and the Pacific in global tomographic
cent accretions was an arc terrane, and there is much simpler geometries, permitting robust pre- images (van der Meer et  al., 2010, 2012; Li
no doubt that they arrived on the Farallon plate. dictions about Mezcalera-Angayucham Oceans et  al., 2008; Obayashi et  al., 2013; Ritsema
suturing by westward subduction, and about the et al., 2011).
5.4. Slab Sinking Rates early transition phase into eastward, Andean-
style Farallon subduction. Sinking rates can be 6. CONCLUSIONS
In sections 2.6 and 2.7, we inferred an aver- robustly estimated as a side product of reconcil-
aged sinking rate for the MEZ-ANG slab walls ing slab wall geometries with plate reconstruc- Paleoreconstruction of oceanic regions, such
of 10 ±  2  mm/yr since subduction, and we tions, and they have been validated against the as the seas west of Mesozoic North America,
discussed how this was not an assumption but geologic record. are plagued by the double uncertainty of where
rather the only rate that is consistent with the ob- the trenches ran and how subducted lithosphere
served slope of the westward-shallowing upper 5.5. Temporal Limitations of our deformed into high-velocity slabs in the mantle.
truncation surfaces of the MEZ and ANG slabs. Archipelago Model—Where Is the Slab Special slab geometries under North America,
Sigloch and Mihalynuk (2013) included the CR that Generated the Triassic–Jurassic showing wall-like linear belts, suggest the
slab wall in this sinking rate estimate. “Native Arc”? null hypothesis that they directly trace out sta-
A lower-mantle sinking rate of 10 ± 2 mm/yr tionary paleotrench locations, and that this
might appear slow compared to other rates sug- For Early Jurassic times, our model agrees lithosphere simply sank vertically beneath the
gested in the literature (see review by Butter- with the Andean analogue as far as (eastward) trenches. Combined with a complete Atlantic
worth et al., 2014). All such estimates have suf- subduction under the native arc (and linked spreading record that allows confident paleo­
fered from uncertainties in trench positions, Inter­montane superterrane) is concerned, sub- positioning of the continent, these slab geom-
possibly quite large and unrecognized, as ar- duction that had shut down by ca. 170 Ma. From etries permit detailed predictions of the exis-
gued here, and from ambiguous matches of reconstructions of the Pangean North American tence, geom­etry, and closure of a major ocean
trenches to slabs. Faster sinking rates tend to margin and from observed systematics of slab ahead of westward-drifting North America, the
arise in convection simulations, but the small sinking, the slab that drove this subduction is Mezcalera-Angayucham Oceans.

1514 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

The most conspicuous prediction for the edge of North America.  This modified margin Mojave-Sonora Megashear Hypothesis: Development,
Assessment, and Alternatives: Geological Society of
geologic record is the presence of this ocean’s proceeded to override the Mezcalera/Angayu- America Special Paper 393, p. 427–455, doi:​10​.1130​/0​
suture, which must run along the entire west- cham arcs—Insular-Guerrero microcontinent -8137​-2393​-0​.427​.
Armstrong, R.L., 1968, Sevier orogenic belt in Nevada and
ern margin of North America, and which we and Alaskan arcs—during a drawn-out collision Utah: Geological Society of America Bulletin, v.  79,
have demonstrated here. Geological evidence from ca. 155 Ma to ca. 50 Ma. All marine geo- p.  429–458, doi:​10​.1130​/0016​-7606​(1968)79​[429:​
confirms the Mezcalera-Angayucham suture physical constraints on Farallon plate evolution SOBINA]2​.0​.CO;2​.
Ash, C.H., 1994, Origin and Tectonic Setting of Ophiolitic
as truly continental in scale, marked by a trail are satisfied by this model. Ultramafic and Related Rocks in the Atlin Area, British
of at least 11 collapsed, Late Jurassic to Late The kind of “tomotectonic” integration dem- Columbia (NTS 104N): British Columbia Ministry of
Cretaceous basins, roughly half of which are onstrated here, with high-resolution tomogra- Energy, Mines and Petroleum Resources Bulletin 94,
48 p.
known to contain mantle rocks (Fig. 4B). They phy, plate reconstructions, and land-geological Asmerom, Y., Zartman, R.E., Damon, P.E., and Shafiqul-
run between the Insular (Peninsular, Alexander, evidence, offers a blueprint for clarifying simi- lah, M., 1990, Zircon U-Th-Pb and whole-rock Rb-Sr
Wrangellia) and Intermontane (Stikinia, Cache age patterns of Lower Mesozoic igneous rocks in the
lar paleogeographic ambiguities in other accre- Santa Rita Mountains, southeast Arizona: Implications
Creek, Quesnellia) microcontinents and were tionary orogens. for Mesozoic magmatism and tectonics in the southern
closed later than the Middle Jurassic or older Cordillera: Geological Society of America Bulletin,
ACKNOWLEDGMENTS v. 102, p. 961–968, doi:​10​.1130​/0016​-7606​(1990)102​
suturing ages invoked by popular versions of the <0961:​ZUTPAW>2​.3​.CO;2​.
Andean-analogue model for the North Ameri- We have benefited from discussions with James Atwater, T., 1970, Implications of plate tectonics for the
can Cordillera. Confirmation of the suture’s pre- Monger, Eldridge Moores, W. Jason Morgan, Rob Cenozoic tectonic evolution of western North Amer-
ica: Geological Society of America Bulletin, v.  81,
dicted location and timing in turn supports the Coe, Edward Mankinen, Stephen Johnston, Terry
p. 3513–3536, doi:​10​.1130​/0016​-7606​(1970)81​[3513:​
hypothesis of vertical slab sinking (although we Poulton, Charles Ferguson, and John Wakabayashi. IOPTFT]2​.0​.CO;2​.
We have also benefited from detailed reviews by Gene Atwater, T., 1989, Plate tectonic history of the Northeast
are not yet able to put tighter bounds on it than Humphreys, Asso­ ciate Editor Clinton Conrad, and Pacific and western North America, in Winterer, E.L.,
Sigloch and Mihalynuk [2013] had obtained two anonymous colleagues. This project has received Hussong, D.M., and Decker, R.W., eds., The Eastern
from the continental deformation record). funding from the European Research Council (ERC) Pacific Ocean and Hawaii: Boulder, Colorado, Geo-
Uncertainty in trench locations has generally under the European Union’s Horizon 2020 research logical Society of America, The Geology of North
and innovation programme (grant agreement 639003 America, v. N, p. 21–72.
been underestimated and gone largely un­noticed Barth, A.P., Walker, J.D., Wooden, J.L., Riggs, N.R., and
“DEEP TIME”). K.S. acknowledges additional fund-
by the deep subsurface community. They have ing from a Philip Leverhulme Prize awarded by The
­Schweickert, R.A., 2011, Birth of the Sierra Nevada
magmatic arc: Early Mesozoic plutonism and volcanism
stuck to the perceived certitude of a single, Leverhulme Trust. in the east-central Sierra Nevada of California: Geo-
margin-­hugging Farallon trench since 180+ Ma, sphere, v. 7, p. 877–897, doi:​10​.1130​/GES00661​.1​.
and eastward subduction under the tightly bound REFERENCES CITED Beaumont, C., Quinlan, G.M., and Stockmal, G.S., 1993,
The evolution of the Western Interior Basin; causes,
package of Insular superterrane, Intermontane Amato, J.M., and Pavlis, T.L., 2010, Detrital zircon ages consequences and unsolved problems, in Caldwell,
superterrane, and cratonic North America, even from the Chugach terrane, southern Alaska, reveal W.G.E., and Kauffman, E.G., eds., Evolution of the
though some geologists had considered alterna- multiple episodes of accretion and erosion in a subduc- Western Interior Basin: Geological Association of
tion complex: Geology, v. 38, p. 459–462, doi:​10​.1130​ Canada Special Paper 39, p.  97–118, https://​dalspace​
tives such as archipelago paleogeographies and /G30719​.1​. .library​.dal​.ca​/handle​/10222​/26207 (accessed October
ribbon continents. Against this disconnect, our Amato, J.M., Bogar, M.J., Gehrels, G.E., Farmer, G.L., 2016).
and McIntosh, W.C., 2007a, The Tlikakila complex Bellahsen, N., Faccenna, C., and Funiciello, F., 2005, Dy-
finding of a large, “missed” paleo-ocean and in southern Alaska: A suprasubduction-zone ophiolite namics of subduction and plate motion in laboratory
its margin-spanning suture is not as unlikely it between the Wrangellia composite terrane and North experiments: Insights into the “plate tectonics” behav-
might sound. Also, while geophysical imaging America, in Ridgway, K.D., Trop, J.M., Glen, J.M.G., ior of the Earth: Journal of Geophysical Research, Solid
and O’Neill, J.M., eds., Tectonic Growth of a Colli- Earth, v. 110, B01401, doi:​10​.1029​/2004JB002999​.
most readily resolves large-scale features such sional Continental Margin: Crustal Evolution of South- Beranek, L.P., Van Staal, C.R., Gordee, S.M., McClelland,
as giant slab walls, this is not the natural length ern Alaska: Geological Society of America Special W.C., Israel, S., and Mihalynuk, M., 2012, Tectonic
scale for geological field work. In the absence Paper 431, p. 227–252, doi:​10​.1130​/2007​.2431​(10)​. significance of Upper Cambrian–Middle Ordovician
Amato, J.M., Rioux, M.E., Kelemen, P.B., Gehrels, G.E., mafic volcanic rocks on the Alexander terrane, Saint
of a guiding hypothesis, there was little reason Clift, P.D., Pavlis, T.L., and Draut, A.E., 2007b, U-Pb Elias Mountains, northwestern Canada: The Journal of
to investigate a dozen Jurassic–Cretaceous ba- geochronology of volcanic rocks from the Jurassic Tal- Geology, v. 120, p. 293–314, doi:​10​.1086​/664788​.
sins from Mexico to Alaska, most of which had keetna Formation and detrital zircons from prearc and Berg, H.C., Jones, D.L., and Richter, D.H., 1972, Gravina-
postarc sequences: Implications for the age of magma- Nutzotin belt—Tectonic significance of an Upper
been displaced along the margin by hundreds to tism and inheritance in the Talkeetna arc, in Ridgway, Meso­zoic sedimentary and volcanic sequence in south-
thousands of kilometers, as a potentially contig- K.D., Trop, J.M., Glen, J.M.G., and O’Neill, J.M., eds., ern and southeastern Alaska, in Geological Survey
Tectonic Growth of a Collisional Continental Margin: Research 1972, Chapter D: U.S. Geological Survey
uous feature, although some studies have come Crustal Evolution of Southern Alaska: Geological So- Professional Paper 800-D, p.  D1–D24, http://​ www​
close (e.g., McClelland et al., 1992). ciety of America Special Paper 431, p. 253–271, doi:​10​ .dggs​.alaska​.gov​/pubs​/id​/4057 (accessed March 2016).
Our slab-guided reconstruction of Cordi­lleran .1130​/2007​.2431​(11)​. Bezada, M.J., Humphreys, E.D., Toomey, D.R., Harnafi, M.,
Amato, J.M., Pavlis, T.L., Clift, P.D., Kochelek, E.J., Dávila, J.M., and Gallart, J., 2013, Evidence for slab
evolution is most similar to geologically derived Hecker, J.P., Worthman, C.M., and Day, E.M., 2013, rollback in westernmost Mediterranean from improved
archipelago models (e.g., Moores, 1970, 1998; Architecture of the Chugach accretionary complex as upper mantle imaging: Earth and Planetary Science
Monger et  al., 1982). It differs from ribbon revealed by detrital zircon ages and lithologic varia- Letters, v.  368, p.  51–60, doi:​10​.1016​/j​.epsl​.2013​.02​
tions: Evidence for Mesozoic subduction erosion in .024​.
continent models (Chamberlain and Lambert, south-central Alaska: Geological Society of America Bijwaard, H., Spakman, W., and Engdahl, E.R., 1998, Clos-
1985; Johnston, 2008; Hildebrand, 2009) in Bulletin, v. 125, p. 1891–1911, doi:​10​.1130​/B30818​.1​. ing the gap between regional and global travel time
Anderson, T.H., and Mahoney, J.B., 2006, Late Jurassic ba- tomography: Journal of Geophysical Research, ser.
that we place no pericratonic terranes of ac- B, Solid Earth, v. 103, p. 30,055–30,078, doi:​10​.1029​
sins; the tracks of a continental-scale transform from
knowledged North American derivation in the the Gulf of Mexico to Alaska, in Haggart, J.W., Enkin, /98JB02467​.
archipelago. We stress that the southeast Inter­ R.J., and Monger, J.W.H., eds., Paleogeography of the Bilodeau, W.L., and Keith, S.B., 1986, Lower Jurassic
North American Cordillera; Evidence For and Against Navajo-Aztec–equivalent sandstones in southern Ari-
montane superterrane was rooted in the cra- Large-Scale Displacements: Geological Association of zona and their paleogeographic significance: American
tonic southwest United States along with coeval Canada Special Paper 46, p. 369–375. Association of Petroleum Geologists Bulletin, v.  70,
Triassic–Early Jurassic “native arcs,” and that Anderson, T.H., Jones, N.W., and McKee, J.W., 2005, The p. 690–701.
Taray Formation; Jurassic(?) mélange in northern Blodgett, R.B., Rohr, D.M., and Boucot, A.J., 2002, Paleo-
those native and allied accreted arcs had col- Mexico; tectonic implications, in Anderson, T.H., zoic links among some Alaskan accreted terranes and
lapsed by 170 Ma to form the then-new leading Nourse, J.A., McKee, J.W., and Steiner, M.B., eds., The Siberia based on megafossils, in Miller, E.L., Grantz,

Geological Society of America Bulletin, v. 129, no. 11/12 1515

 Sigloch and Mihalynuk

A., and Klemperer, S., eds., Tectonic Evolution of the Butler, R.F., Gehrels, G.E., Hart, W., Davidson, C., and 10 ​ . 1130 ​ / 0091 ​ - 7613 ​ ( 1993)021 ​ < 0263: ​ L LPRTB>2​ . 3​
Bering Shelf–Chukchi Sea–Arctic Margin and Adjacent Crawford, M.L., 2006, Paleomagnetism of Late Juras- .CO;2​.
Areas: Geological Society of America Special ­Paper sic to mid-Cretaceous plutons near Prince Rupert, Brit- Cordey, F., Mortimer, N., DeWever, P., and Monger, J.W.H.,
360, p. 273–290, doi:​10​.1130​/0​-8137​-2360​-4​.273​. ish Columbia, in Haggart, J., Enkin, R.J., and Monger, 1987, Significance of Jurassic radiolarians from the
Box, S.E., and Patton, W.W., 1989, Igneous history of J.W.H., eds., Paleogeography of the North American Cache Creek terrane, British Columbia: Geology,
the Koyukuk terrane, western Alaska; constraints Cordillera: Evidence For and Against Large-Scale Dis- v. 15, p. 1151–1154, doi:​10​.1130​/0091​-7613​(1987)15​
on the origin, evolution, and ultimate collision of an placements: Geological Association of Canada Special <1151:​SOJRFT>2​.0​.CO;2​.
accreted island arc terrane: Journal of Geophysi- Paper 46, p. 171–200. Cowan, D.S., 2003, Revisiting the Baranof–Leech River
cal Research, v.  94, p.  15,843–15,867, doi:​ 10​.1029​ Butterworth, N.P., Talsma, A.S., Müller, R.D., Seton, M., hypothesis for Early Tertiary coastwise transport of
/JB094iB11p15843​. Bunge, H.P., Schuberth, B.S.A., Shephard, G.E., and the Chugach–Prince William terrane: Earth and Plan-
Bradley, D.C., McClelland, W.C., Wooden, J.L., Till, A.B., Heine, C., 2014, Geological, tomographic, kinematic etary Science Letters, v. 213, p. 463–475, doi:​10​.1016​
Roeske, S.M., Miller, M.L., Karl, S.M., and Abbott, and geodynamic constraints on the dynamics of sink- /S0012​-821X​(03)00300​-5​.
J.G., 2007, Detrital zircon geochronology of some ing slabs: Journal of Geodynamics, v. 73, p. 1–13, doi:​ Crawford, M.L., Hollister, L.S., and Woodsworth, G.J.,
Neoproterozoic to Triassic rocks in interior Alaska, in 10​.1016​/j​.jog​.2013​.10​.006​. 1987, Crustal deformation and regional metamorphism
Dumoulin, J.A., and Till, A.B., eds., Late Proterozoic Cecil, M.R., Rotberg, G.L., Ducea, M.N., Saleeby, J.B., across a terrane boundary, Coast plutonic complex,
to Devonian Continental Sequence, Alaska: Geological and Gehrels, G.E., 2012, Magmatic growth and batho- British Columbia: Tectonics, v. 6, p. 343–361, doi:​10​
Society of America Special Paper 431, p. 155–189, doi:​ lithic root development in the northern Sierra Nevada, .1029​/TC006i003p00343​.
10​.1130​/2007​.2431​(07)​. California: Geosphere, v.  8, p.  592–606, doi:​10​.1130​ Crawford, M.L., Klepeis, K.A., Gehrels, G.E., and Lind-
Brandon, M.T., 1989, Deformational styles in a sequence of /GES00729​.1​. line, J., 2009, Mid-Cretaceous–Recent crustal evolu-
olistostromal mélanges, Pacific Rim complex, western Centeno-Garcia, E., Busby, C., Busby, M., and Gehrels, G., tion in the central Coast orogen, British Columbia and
Vancouver Island, Canada (with Suppl. Data 89–16): 2011, Evolution of the Guerrero composite terrane southeastern Alaska, in Miller, R.B., and Snoke, A.W.,
Geological Society of America Bulletin, v.  101, along the Mexican margin, from extensional fringing eds., Crustal Cross Sections from the Western North
p.  1520–1542, doi:​1 0​. 1130​/ 0016​- 7606​( 1989)101​ arc to contractional continental arc: Geological Society American Cordillera and Elsewhere: Implications for
<1520:​DSIASO>2​.3​.CO;2​. of America Bulletin, v. 123, p. 1776–1797, doi:​10​.1130​ Tectonic and Petrologic Processes: Geological Society
Brew, D.A., and Karl, S.M., 1988, A reexamination of the /B30057​.1​. of America Special Paper 456, p. 97–124, doi:​10​.1130​
contacts and other features of the Gravina belt, south- Chamberlain, V.E., and Lambert, R.S.J., 1985, Cordilleria, /2009​.2456​(04)​.
eastern Alaska, in Galloway, J.P., and Hamilton, T.D., a newly defined Canadian microcontinent: Nature, Csejtey, B., Cox, D.P., Evarts, R.C., Stricker, G.D., and Fos-
eds., Geologic Studies in Alaska by the U.S. Geologi- v. 314, p. 707–713, doi:​10​.1038​/314707a0​. ter, H.L., 1982, The Cenozoic Denali fault system and
cal Survey during 1987: U.S. Geological Survey Circu- Christiansen, E.H., Kowallis, B.J., and Barton, M.D., 1994, the Cretaceous accretionary development of southern
lar 1016, p. 143–146. Temporal and spatial distribution of volcanic ash in Alaska: Journal of Geophysical Research, ser. B, v. 87,
Brown, B.J., Müller, R.D., Gaina, C., Struckmeyer, H.I.M., Mesozoic sedimentary rocks of the Western Interior: Special Issue on Accretion Tectonics, p.  3741–3754,
Stagg, H.M.J., and Symonds, P.A., 2003, Formation An alternative record of Mesozoic magmatism, in doi:​10​.1029​/JB087iB05p03741​.
and evolution of Australian passive margins: Implica- ­Caputo, M.V., Peterson, J.A., and Franczyk, K.J., eds., Cutts, J.A., McNicoll, V.J., Zagorevski, A., Anderson, R.G.,
tions for locating the boundary between continental Mesozoic Systems of the Rocky Mountain Region, and Martin, K., 2015, U-Pb geochronology of the
and oceanic crust, in Hillis, R.R., and Müller, R.D., United States: Denver, Colorado, Rocky Mountain Hazel­ton Group in the McTagg anticlinorium, Iskut
eds., Evolution and Dynamics of the Australian Plate: Section, Society for Sedimentary Geology (SEPM), River area, northwestern British Columbia, in Geologi-
Geological Society of America Special Paper  372, p. 73–94. cal Fieldwork 2014: British Columbia Geological Sur-
p. 223–243, doi:​10​.1130​/0​-8137​-2372​-8​.223​. Čížková, H., and Bina, C.R., 2013, Effects of mantle and vey Paper 2015-1, p. 87–101, http://​www​.em​.gov​.bc​.ca​
Brown, E.H., 2012, Obducted nappe sequence in the San subduction-interface rheologies on slab stagnation and /Mining​/Geoscience​/PublicationsCatalogue​/Fieldwork​
Juan Islands–northwest Cascades thrust system, Wash- trench rollback: Earth and Planetary Science Letters, /Documents​/2014​/05​_Cutts​_etal​.pdf.
ington and British Columbia: Canadian Journal of Earth v. 379, p. 95–103, doi:​10​.1016​/j​.epsl​.2013​.08​.011​. Dahlen, F.A., Hung, S.-H., and Nolet, G., 2000, Fréchet ker-
Sciences, v. 49, p. 796–817, doi:​10​.1139​/e2012​-026​. Clift, P.D., Wares, N.M., Amato, J.M., Pavlis, T.L., Hole, nels for finite-frequency traveltimes—I: Theory: Geo-
Brown, E.H., Lapen, T.J., Leckie, R.M., Silva, I.P., Verga, M.J., Worthman, C., and Day, E., 2012, Evolving physical Journal International, v. 141, p. 157–174, doi:​
D., and Singer, B.S., 2005, Revised ages of blueschist heavy mineral assemblages reveal changing exhuma- 10​.1046​/j​.1365​-246X​.2000​.00070​.x​.
metamorphism and the youngest pre-thrusting rocks in tion and trench tectonics in the Mesozoic Chugach ac- Davidson, C., and McPhillips, D., 2007, Along strike varia-
the San Juan Islands, Washington: Canadian Journal of cretionary complex, south-central Alaska: Geological tions in metamorphism and deformation of the strata of
Earth Sciences, v. 42, p. 1389–1400, doi:​10​.1139​/e05​ Society of America Bulletin, v. 124, p. 989–1006, doi:​ the Kahiltna basin, south-central Alaska, in Ridgway,
-033​. 10​.1130​/B30594​.1​. K.D., Trop, J.M., Glen, J.M.G., and O’Neill, J.M., eds.,
Bunge, H.-P., and Grand, S.P., 2000, Mesozoic plate-motion Cloos, M., Sapiie, B., van Ufford, A.Q., Weiland, R.J., Tectonic Growth of a Collisional Continental Margin:
history below the northeast Pacific Ocean from seismic ­Warren, P.Q., and McMahon, T.P., 2005, Collisional Crustal Evolution of Southern Alaska: Geological So-
images of the subducted Farallon slab: Nature, v. 405, delamination in New Guinea: The geotectonics of sub- ciety of America Special Paper 431, p. 439–453, doi:​10​
p. 337–340, doi:​10​.1038​/35012586​. ducting slab breakoff: Geological Society of America .1130​/2007​.2431​(17)​.
Burchfiel, B.C., and Davis, G.A., 1972, Structural frame- Special Paper 400, 51 p. Day, E.M., Pavlis, T.L., and Amato, J.M., 2011, Diverging
work and evolution of the southern part of the Cor­ Clowes, R.M., Brandon, M.T., Green, A.G., Yorath, C.J., histories of the Liberty Creek and Iceberg Lake blue-
di­lleran orogen, western United States: American Brown, A.S., Kanasewich, E.R., and Spencer, C., schist bodies, south central Alaska: San Francisco,
Journal of Science, v. 272, p. 97–118, doi:​10​.2475​/ajs​ 1987, LITHOPROBE—southern Vancouver Island: California, American Geophysical Union, Fall Meeting
.272​.2​.97​. Cenozoic subduction complex imaged by deep seismic abstracts, v. 1, p. 2426.
Burchfiel, B.C., and Davis, G.A., 1975, Nature and controls reflections: Canadian Journal of Earth Sciences, v. 24, DeCelles, P.G., 2004, Late Jurassic to Eocene evolution of
of Cordilleran orogenesis, western United States—Ex- p. 31–51, doi:​10​.1139​/e87​-004​. the Cordilleran thrust belt and foreland basin system,
tensions of an earlier synthesis: American Journal of Coe, R.S., Globerman, B.R., and Thrupp, G.A., 1989, Rota- western U.S.A.: American Journal of Science, v. 304,
Science, v. A275, p. 363–396. tion of central and southern Alaska in the Early Ter- p. 105–168, doi:​10​.2475​/ajs​.304​.2​.105​.
Burchfiel, B.C., Cowan, D.S., and Davis, G.A., 1992, Tec- tiary: Oroclinal bending by megakinking?, in Kissel, DeCelles, P.G., and Coogan, J.C., 2006, Regional structure
tonic overview of the Cordilleran orogen in the western C., and Laj, C., eds., Paleomagnetic Rotations and and kinematic history of the Sevier fold-and-thrust
United States, in Burchfiel, B.C., Lipman, P.W., and Continental Deformation: Springer, p. 327–342, http://​ belt, central Utah: Geological Society of America Bul-
Zoback, M.L., eds., The Cordilleran Orogen: Conter- link​.springer​.com​/chapter​/10​.1007​/978​-94​-009​-0869​ letin, v. 118, p. 841–864, doi:​10​.1130​/B25759​.1​.
minous U.S.: Boulder, Colorado, Geological Society -7_20 (accessed April 2015), doi:​10​.1007​/978​-94​-009​ Dewey, J.F., 1988, Extensional collapse of orogens: Tectonics,
of America, The Geology of North America, v.  G-3, -0869​-7_20​. v. 7, p. 1123–1139, doi:​10​.1029​/TC007i006p01123​.
p. 407–479. Cohen, K.M., Finney, S.C., Gibbard, P.L., and Fan, J.X., Dickinson, W.R., 1976, Sedimentary basins developed dur-
Burdick, S., Li, C., Martynov, V., Cox, T., Eakins, J., Mul- 2013, The ICS International Chronostratigraphic ing evolution of Mesozoic–Cenozoic arc-trench system
der, T., Astiz, L., Vernon, F.L., Pavlis, G.L., and van Chart: Episodes, v. 36, p. 199–204. in western North America: Canadian Journal of Earth
der Hilst, R.D., 2008, Upper mantle heterogeneity be- Colpron, M., Warren, M.J., and Price, R.A., 1998, Selkirk Sciences, v. 13, p. 1268–1287, doi:​10​.1139​/e76​-129​.
neath North America from travel time tomography with fan structure, southeastern Canadian Cordillera; tec- Dickinson, W.R., 1979, Mesozoic forearc basin in central
global and USArray transportable array data: Seismo- tonic wedging against an inherited basement ramp: Oregon: Geology, v. 7, p. 166–170, doi:​10​.1130​/0091​
logical Research Letters, v. 79, p. 384–392, doi:​10​.1785​ Geological Society of America Bulletin, v.  110, -7613​(1979)7​<166:​MFBICO>2​.0​.CO;2​.
/gssrl​.79​.3​.384​. p.  1060–1074, doi:​1 0​. 1130​/ 0016​- 7606​( 1998)110​ Dickinson, W.R., 1981, Plate tectonic evolution of the south-
Burdick, S., Van der Hilst, R.D., Vernon, F.L., Martynov, V., <1060:​SFSSCC>2​.3​.CO;2​. ern Cordillera: Relations of tectonics to ore deposits
Cox, T., Eakins, J., Karasu, G.H., Tylell, J., Astiz, L., Coney, P.J., Jones, D.L., and Monger, J.W.H., 1980, Cor­di­ in the southern Cordillera: Arizona Geological Society
and Pavlis, G.L., 2014, Model update January 2013: lleran suspect terranes: Nature, v. 288, p. 329–333, doi:​ Digest, v. 14, p. 113–135.
Upper mantle heterogeneity beneath North America 10​.1038​/288329a0​. Dickinson, W.R., 2004, Evolution of the North American
from travel-time tomography with global and USArray Cordey, F., and Schiarizza, P., 1993, Long-lived Pantha­ Cordillera: Annual Review of Earth and Planetary Sci-
transportable array data: Seismological Research Let- lassic remnant; the Bridge River accretionary complex, ences, v.  32, p.  13–45, doi:​10​.1146​/annurev​.earth​.32​
ters, v. 85, p. 77–81, doi:​10​.1785​/0220130098​. Canadian Cordillera: Geology, v. 21, p. 263–266, doi:​ .101802​.120257​.

1516 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

Dickinson, W.R., 2006, Geotectonic evolution of the Great Ernst, W.G., 2011, Accretion of the Franciscan complex at- Gehrels, G., Rusmore, M., Woodsworth, G., Crawford,
Basin: Geosphere, v.  2, p.  353–368, doi:​ 10​.1130​ tending Jurassic–Cretaceous geotectonic development M., Andronicos, C., Hollister, L., Patchett, J., Ducea,
/GES00054​.1​. of northern and central California: Geological Society M., Butler, R., Klepeis, K., Davidson, C., Friedman,
Dickinson, W.R., 2008, Accretionary Mesozoic–Ceno- of America Bulletin, v. 123, p. 1667–1678, doi:​10​.1130​ R., Haggart, J., Mahoney, B., et  al., 2009, U-Th-Pb
zoic expansion of the Cordilleran continental margin /B30398​.1​. geochronology of the Coast Mountains batholith in
in California and adjacent Oregon: Geosphere, v.  4, Ernst, W.G., Saleeby, J.B., and Snow, C.A., 2009, Guada- north-coastal British Columbia: Constraints on age
p. 329–353, doi:​10​.1130​/GES00105​.1​. lupe Pluton–Mariposa Formation age relationships and tectonic evolution: Geological Society of America
Dickinson, W.R., 2013, Phanerozoic palinspastic recon- in the southern Sierran Foothills; onset of Meso- Bulletin, v. 121, p. 1341–1361, doi:​10​.1130​/B26404​.1​.
structions of Great Basin geotectonics (Nevada-Utah, zoic subduction in Northern California?: Journal of Gibert, G., Gerbault, M., Hassani, R., and Tric, E., 2012,
USA): Geosphere, v.  9, p.  1384–1396, doi:​10​.1130​ Geophysical Research, v.  114, B11204, doi:​10​.1029​ Dependency of slab geometry on absolute velocities
/GES00888​.1​. /2009JB006607​. and conditions for cyclicity: Insights from numerical
Dickinson, W.R., and Gehrels, G.E., 2009, U-Pb ages of de­ Evenchick, C.A., and McNicoll, V.J., 2002, Stratigraphy, modelling: Geophysical Journal International, v.  189,
trital zircons in Jurassic eolian and associated sandstones structure, and geochronology of the Anyox Pendant, p. 747–760, doi:​10​.1111​/j​.1365​-246X​.2012​.05426​.x​.
of the Colorado Plateau: Evidence for transcontinental northwest British Columbia, and implications for min- Goldhammer, R.K., and Johnson, C.A., 1999, Mesozoic
dispersal and intraregional recycling of sediment: Geo- eral exploration: Canadian Journal of Earth Sciences, sequence stratigraphy and paleogeographic evolution
logical Society of America Bulletin, v. 121, p. 408–433, v. 39, p. 1313–1332, doi:​10​.1139​/e02​-036​. of northeast Mexico, in Bartolini, C., Wilson, J.L.,
doi:​10​.1130​/B26406​.1​. Evenchick, C.A., McMechan, M.E., McNicoll, V.J., and and Lawton, T.F., eds., Mesozoic Sedimentary and
Dickinson, W.R., and Gehrels, G.E., 2010, Insights into Carr, S.D., 2007, A synthesis of the Jurassic–Creta- Tectonic History of North-Central Mexico: Geological
North American paleogeography and paleotectonics ceous tectonic evolution of the central and southeast- Society of America Special Paper 340, p. 1–58, doi:​10​
from U-Pb ages of detrital zircons in Mesozoic strata ern Canadian Cordillera: Exploring links across the .1130​/0​-8137​-2340​-X​.1​.
of the Colorado Plateau, USA: International Journal orogen, in Sears, J.W., Harms, T.A., and Evenchick, Grand, S.P., 1994, Mantle shear structure beneath the Amer-
of Earth Sciences, v.  99, p.  1247–1265, doi:​10​.1007​ C.A., eds., Whence the Mountains? Inquiries into the icas and surrounding oceans: Journal of Geophysi-
/s00531​-009​-0462​-0​. Evolution of Orogenic Systems: A Volume in Honor cal Research, v.  99, p.  11,591–11,621, doi:​10​.1029​
Dickinson, W.R., and Lawton, T.F., 2001, Carboniferous of Raymond A. Price: Geological Society of America /94JB00042​.
to Cretaceous assembly and fragmentation of Mex- Special Paper 433, p. 117–145, doi:​10​.1130​/2007​.2433​ Grand, S.P., van der Hilst, R.D., and Widiyantoro, S., 1997,
ico: Geological Society of America Bulletin, v.  113, (06)​. Global seismic tomography; a snapshot of convection
p. 1142–1160, doi:​10​.1130​/0016​-7606​(2001)113​<1142:​ Evenchick, C.A., Poulton, T.P., and McNicoll, V.J., 2010, in the Earth: GSA Today, v. 7, no. 4, p. 1–7.
CTCAAF>2​.0​.CO;2​. Nature and significance of the diachronous contact Greene, A.R., Scoates, J.S., Weis, D., Katvala, E.C., Israel,
Dickinson, W.R., Gehrels, G.E., and Stern, R.J., 2010, Late between the Hazelton and Bowser Lake Groups (Ju- S., and Nixon, G.T., 2010, The architecture of oceanic
Triassic Texas uplift preceding Jurassic opening of the rassic), north-central British Columbia: Bulletin of plateaus revealed by the volcanic stratigraphy of the
Gulf of Mexico: Evidence from U-Pb ages of de­trital Canadian Petroleum Geology, v. 58, p. 235–267, doi:​ accreted Wrangellia oceanic plateau: Geosphere, v. 6,
zircons: Geosphere, v.  6, p.  641–662, doi:​10​.1130​ 10​.2113​/gscpgbull​.58​.3​.235​. p. 47–73, doi:​10​.1130​/GES00212​.1​.
/GES00532​.1​. Friedman, R.M., and Armstrong, R.L., 1995, Jurassic and Groome, W.G., Thorkelson, D.J., Friedman, R.M.,
du Bray, E.A., John, D.A., and Cousens, B.L., 2014, Petro­ Cretaceous geochronology of the Southern Coast belt, Mortensen, J.K., Massey, N.W., Marshall, D.D., and
logic, tectonic, and metallogenic evolution of the British Columbia, 49° to 51°N, in Miller, D.M., and Layer, P.W., 2003, Magmatic and tectonic history of the
southern segment of the ancestral Cascades magmatic Busby, C., eds., Jurassic Magmatism and Tectonics of Leech River complex, Vancouver Island, British Co-
arc, California and Nevada: Geosphere, v. 10, p. 1–39, the North American Cordillera: Geological Society of lumbia: Evidence for ridge-trench intersection and ac-
doi:​10​.1130​/GES00944​.1​. America Special Paper 299, p.  95–140, doi:​10​.1130​ cretion of the Crescent terrane, in Sisson, V.B., Roeske,
Ducea, M., 2001, The California arc: Thick granitic batho- /SPE299​-p95​. S.M., and Pavlis, T.L., eds., Geology of a Transpres-
liths, eclogitic residues, lithospheric-scale thrusting, Fukao, Y., Obayashi, M., and Nakakuki, T., 2009, Stagnant sional Orogen Developed During Ridge-Trench Inter­
and magmatic flare-ups: GSA Today, v.  11, p.  4–10, slab: A review: Annual Review of Earth and Planetary action along the North Pacific Margin: Geological
doi:​10​.1130​/1052​-5173​(2001)011​<0004:​TCATGB>2​ Sciences, v.  37, p.  19–46, doi:​10​.1146​/annurev​.earth​ Society of America Special Paper 371, p. 327–353, doi:​
.0​.CO;2​. .36​.031207​.124224​. 10​.1130​/0​-8137​-2371​-X​.327​.
Dumitru, T.A., Wakabayashi, J., Wright, J.E., and Wooden, Funiciello, F., Faccenna, C., Giardini, D., and Regenauer- Grove, M., Lovera. O., and Harrison. M., 2003, Late Cre-
J.L., 2010, Early Cretaceous transition from non­accre­ Lieb, K., 2003, Dynamics of retreating slabs: 2. In- taceous cooling of the east-central Peninsular Ranges
tionary behavior to strongly accretionary behavior sights from three-dimensional laboratory experiments: batholith (33°N): Relationship to La Posta pluton
within the Franciscan subduction complex: Tectonics, Journal of Geophysical Research–Solid Earth, v. 108, emplacement, Laramide shallow subduction, and
v. 29, TC5001, doi:​10​.1029​/2009TC002542​. no. B4, 2207, doi:​10​.1029​/2001JB000896​. forearc sedimentation, in Johnson, S.E., Paterson,
Duncan, R.A., and Richards, M.A., 1991, Hotspots, man- Gardner, M.C., Bergman, S.C., Cushing, G.W., MacKevett, S.R., Fletcher, J.M.. Girty, G.H., Kimbrough, D.L., and
tle plumes, flood basalts, and true polar wander: Re- E.M., Plafker, G., Campbell, R.B., Dodds, C.J., Mc- Martin-Barajas, A., eds., Tectonic Evolution of North-
views of Geophysics, v.  29, p.  31–50, doi:​ 10​.1029​ Clelland, W.C., and Mueller, P.A., 1988, Pennsyl­vanian western Mexico and the Southwestern USA: Boulder,
/90RG02372​. pluton stitching of Wrangellia and the Alexander ter- Colorado, Geological Society of America Special P ­ aper
Dusel-Bacon, C., Lanphere, M.A., Sharp, W.D., Layer, P.W., rane, Wrangell Mountains, Alaska: Geology, v. 16, 374, p. 355–379 doi:​10​.1130​/0​-8137​-2374​-4​.355​.
and Hansen, V.L., 2002, Mesozoic thermal history p.  967–971, doi:​10​.1130​/0091​-7613​(1988)016​<0967:​ Guillou-Frottier, L., Buttles, J., and Olson, P., 1995, Labora-
and timing of structural events for the Yukon-Tanana PPSOWA>2​.3​.CO;2​. tory experiments on the structure of subducted litho-
Upland, east-central Alaska: 40Ar/39Ar data from meta- Garel, F., Goes, S., Davies, D.R., Davies, J.H., Kramer, S.C., sphere: Earth and Planetary Science Letters, v.  133,
morphic and plutonic rocks: Canadian Journal of Earth and Wilson, C.R., 2014, Interaction of subducted slabs p. 19–34, doi:​10​.1016​/0012​-821X​(95)00045​-E​.
Sciences, v. 39, p. 1013–1051, doi:​10​.1139​/e02​-018​. with the mantle transition-zone: A regime diagram Haggart, J.W., Mahoney, J.B., Forgette, M., Carter, E.S.,
Edgar, N.T., 2003, A modern analogue for tectonic, eustatic, from 2-D thermo-mechanical models with a mobile Schröder-Adams, C.J., MacLaurin, C.I., and Sweet,
and climatic processes in cratonic basins: Gulf of Car- trench and an overriding plate: Geochemistry Geo- A.R., 2011, Paleoenvironmental and chronological
penteria, northern Australia, in Cecil, C.B., and Edgar, physics Geosystems, v. 15, p. 1739–1765, doi:​10​.1002​ constraints on the Mount Tatlow succession, British
N.T., eds., Climate Controls on Stratigraphy: Society /2014GC005257​. Columbia: First recognition of radiolarian and forami-
for Sedimentary Geology (SEPM) Special Publication Garver, J.I., 1992, Provenance of Albian–Cenomanian rocks niferal faunas in the Intermontane Cretaceous back-arc
77, http://​archives​.datapages​.com​/data​/sepm​_sp​/SP77​ of the Methow and Tyaughton Basins, southern Brit- basins of western Canada: Canadian Journal of Earth
/A​ _ Modern​ _ Analogue​ _ for ​ _ Tectonic ​ _ Eustatic ​ _ and​ ish Columbia; a mid-Cretaceous link between North Sciences, v. 48, p. 952–972, doi:​10​.1139​/e11​-019​.
_Climatic​_Processes​.pdf (accessed October 2016). America and the Insular terrane: Canadian Journal of Hamilton, W.B., 1969, Mesozoic California and the under-
Engebretson, D.C., Cox, A., and Gordon, R.G., 1985, Rela- Earth Sciences, v. 29, p. 1274–1295, doi:​10​.1139​/e92​ flow of Pacific mantle: Geological Society of America
tive Motions Between Oceanic and Continental Plates -102​. Bulletin, v. 80, p. 2409–2430, doi:​10​.1130​/0016​-7606​
in the Pacific Basin: Geological Society of Amer- Gaschnig, R.M., Vervoort, J.D., Lewis, R.S., and Tikoff, B., (1969)80​[2409:​MCATUO]2​.0​.CO;2​.
ica. Special Paper 206, 60 p., doi:​10​.1130​/SPE206​-p1​. 2011, Isotopic evolution of the Idaho batholith and Hamilton, W.B., 1979, Tectonics of the Indonesian Re-
Enkin, R.J., 2006, Paleomagnetism and the case for Baja– Challis intrusive province, northern US Cordillera: gion: U.S. Geological Survey Professional Paper
British Columbia: Paleogeography of the North Ameri- Journal of Petrology, v. 52, p. 2397–2429, doi:​10​.1093​ 1078, 345  p., https://​p ubs​. er​. usgs​. gov​/ publication​
can Cordillera, in Haggart, J., Enkin, R.J., and Monger, /petrology​/egr050​. /pp1078 (accessed October 2016).
J.W.H., eds., Paleogeography of the North American Gehrels, G.E., 2001, Geology of the Chatham Sound region, Hampton, B.A., Ridgway, K.D., and Gehrels, G.E., 2007,
Cordillera: Evidence For and Against Large-Scale Dis- southeast Alaska and coastal British Columbia: Cana- U-Pb detrital zircon age data from Upper Triassic strata
placements: Geological Association of Canada Special dian Journal of Earth Sciences, v.  38, p.  1579–1599, of the Chulitna terrane, south-central Alaska: A prelim-
Paper 46, p. 233–253. doi:​10​.1139​/e01​-040​. inary link with the western margin of North America:
Ernst, W.G., 1970, Tectonic contact between the Franciscan Gehrels, G.E., Butler, R.F., and Bazard, D.R., 1996, ­Detrital Geological Society of America Abstracts with Pro-
mélange and the Great Valley sequence; crustal ex- zircon geochronology of the Alexander terrane, south- grams, v.  39, no.  4, p.  69 https://​gsa​.confex​.com​/gsa​
pression of a late Mesozoic Benioff zone: Journal of eastern Alaska: Geological Society of America Bul- /2007CD​/finalprogram​/abstract​_121407​.htm.
Geophysical Research, v. 75, p. 886–901, doi:​10​.1029​ letin, v.  108, p.  722–734, doi:​10​.1130​/0016​-7606​ Hampton, B.A., Ridgway, K.D., and Gehrels, G.E., 2010,
/JB075i005p00886​. (1996)108​<0722:​DZGOTA>2​.3​.CO;2​. A detrital record of Mesozoic island arc accretion and

Geological Society of America Bulletin, v. 129, no. 11/12 1517

 Sigloch and Mihalynuk

exhumation in the North American Cordillera: U-Pb p.  10042, http://​adsabs​.harvard​.edu​/abs​/2015EGUGA​ lision in an accretionary orogen: Scientific Reports,
geochronology of the Kahiltna basin, southern Alaska: .1710042J (accessed April 2017). v. 6, p. 1–12 http://​www​.ncbi​.nlm​.nih​.gov​/pmc​/articles​
Tectonics, v. 29, TC4015, doi:​10​.1029​/2009TC002544​. Johnston, S.T., 2001, The Great Alaskan terrane wreck: /PMC4863176/ (accessed November 2016).
Harper, G.D., Saleeby, J.B., and Heizler, M., 1994, For- Reconciliation of paleomagnetic and geological data Lithgow-Bertelloni, C., and Richards, M.A., 1998, The
mation and emplacement of the Josephine ophiolite in the northern Cordillera: Earth and Planetary Science dynamics of Cenozoic and Mesozoic plate motions:
and the Nevadan orogeny in the Klamath Mountains, Letters, v. 193, p. 259–272, doi:​10​.1016​/S0012​-821X​ Reviews of Geophysics, v.  36, p.  27–78, doi:​10​.1029​
California-Oregon: U/Pb zircon and 40Ar/39Ar geochro- (01)00516​-7​. /97RG02282​.
nology: Journal of Geophysical Research–Solid Earth, Johnston, S.T., 2008, The Cordilleran ribbon continent of Liu, L., 2014, Constraining Cretaceous subduction polarity
v. 99, p. 4293–4321, doi:​10​.1029​/93JB02061​. North America: Annual Review of Earth and Planetary in eastern Pacific from seismic tomography and geo­
Hildebrand, R.S., 2009, Did Westward Subduction Cause Sciences, v. 36, p. 495–530, doi:​10​.1146​/annurev​.earth​ dynamic modeling: Geophysical Research Letters,
Cretaceous–Tertiary Orogeny in the North American .36​.031207​.124331​. v. 89, p. 221–229.
Cordillera?: Geological Society of America Special Jones, D.L., Silberling, N.J., Gilbert, W., and Coney, P., Liu, L., Spasojević, S., and Gurnis, M., 2008, Reconstruct-
Paper 457, 71 p., doi:​10​.1130​/2009​.2457​. 1982, Character, distribution, and tectonic significance ing Farallon plate subduction beneath North America
Hildebrand, R.S., 2012, Mesozoic Assembly of the of accretionary terranes in the central Alaska Range: back to the Late Cretaceous: Science, v. 322, p. 934–
North American Cordillera: Geological Society of Journal of Geophysical Research–Solid Earth, v.  87, 938, doi:​10​.1126​/science​.1162921​.
America Special Paper 495, 182  p., doi:​ 10​
.1130​ p. 3709–3717, doi:​10​.1029​/JB087iB05p03709​. Liu, L., Gurnis, M., Seton, M., Saleeby, J., Müller, R.D., and
/9780813724959​. Journeay, J.M., and Friedman, R.M., 1993, The Coast belt Jackson, J.M., 2010, The role of oceanic plateau sub-
Hildebrand, R.S., 2015, Dismemberment and northward thrust system; evidence of Late Cretaceous shorten- duction in the Laramide orogeny: Nature Geoscience,
migration of the Cordilleran orogen: Baja-BC re- ing in southwest British Columbia: Tectonics, v.  12, v. 3, p. 353–357, doi:​10​.1038​/ngeo829​.
solved: GSA Today, v. 25, no. 11, p. 4–11, http://​rock​ p. 756–775, doi:​10​.1029​/92TC02773​. Livaccari, R.F., Burke, K., and Şengör, A.M.C., 1981, Was
.geosociety​.org​/gsatoday​/archive​/25​/11​/article​/i1052​ Journeay, J.M., and Mahoney, J.B., 1994, Cayoosh as- the Laramide orogeny related to subduction of an oce-
-5173​-25​-11​-4​.htm. semblage: Regional correlations and implications for anic plateau?: Nature, v. 289, p. 276–278, doi:​10​.1038​
Hillhouse, J.W., and Coe, R.S., 1994, Paleomagnetic data terrane linkages in the southern Coast belt, British /289276a0​.
from Alaska, in Plafker, G., and Berg, H.C., eds., The Columbia: Geological Survey of Canada Current Re- Logan, J.M., Drobe, J.R., and McClelland, W.C., 2000,
Geology of Alaska: Boulder, Colorado, Geological search 1994-A, p. 165–175. Geol­ogy of the Forest Kerr–Mess Creek Area, North-
Society of America, the Geology of North America, Journeay, J.M., and Northcote, B.R., 1992, Tectonic as- western British Columbia: British Columbia Ministry
v. G-1, p. 797–812. semblages of the eastern Coast belt, southwest British of Energy and Mines Bulletin 104, 164 p.
Himmelberg, G.R., Haeussler, P.J., and Brew, D.A., 2004, Columbia: Geological Survey of Canada Current Re- López-Carmona, A., Kusky, T.M., Santosh, M., and Abati,
Emplacement, rapid burial, and exhumation of 90-Ma search Paper 92-1A, p. 215–224. J., 2011, P-T and structural constraints of lawsonite
plutons in southeastern Alaska: Canadian Journal of Kalbas, J.L., Ridgway, K.D., and Gehrels, G.E., 2007, Stra- and epidote blueschists from Liberty Creek and Sel-
Earth Sciences, v.  41, p.  87–102, doi:​ 10​.1139​/e03​ tigraphy, depositional systems, and provenance of the dovia: Tectonic implications for early stages of sub-
-087​. Lower Cretaceous Kahiltna assemblage, western Alaska duction along the southern Alaska convergent margin:
Hollister, L.S., and Crawford, M.L., 1986, Melt-enhanced Range: Basin development in response to oblique colli- Lithos, v.  121, p.  100–116, doi:​10​.1016​/j​.lithos​.2010​
deformation: A major tectonic process: Geology, v. 14, sion, in Ridgway, K.D., Trop, J.M., Glen, J.M.G., and .10​.007​.
p.  558–561, doi:​10​.1130​/0091​-7613​(1986)14​<558:​ O’Neill, J.M., eds., Tectonic Growth of a Collisional MacIntyre, D.G., Villeneuve, M.E., and Schiarizza, P.,
MDAMTP>2​.0​.CO;2​. Continental Margin: Crustal Evolution of South- 2001, Timing and tectonic setting of Stikine terrane
Hosseini, K., and Sigloch, K., 2015, Multifrequency mea- ern Alaska: Geological Society of America ­Special
magmatism, Babine-Takla Lakes area, central British
surements of core-diffracted P waves (Pdiff) for global ­Paper 431, p. 307–343, doi:​10​.1130​/2007​.2431​(13)​.
Columbia: Canadian Journal of Earth Sciences, v. 38,
waveform tomography: Geophysical Journal Interna- Kennett, B.L.N., and Engdahl, E.R., 1991, Traveltimes for
p. 579–601, doi:​10​.1139​/e00​-105​.
tional, v. 203, p. 506–521, doi:​10​.1093​/gji​/ggv298​. global earthquake location and phase identification:
Mahoney, J.B., and Journeay, J.M., 1993, The Cayoosh as-
Houseman, G.A., and Molnar, P., 1997, Gravitational (Ray- Geophysical Journal International, v. 105, p. 429–465,
semblage, southwestern British Columbia: Last vestige
leigh–Taylor) instability of a layer with non-linear doi:​10​.1111​/j​.1365​-246X​.1991​.tb06724​.x​.
of the Bridge River Ocean, in Current Research, Part A,
viscosity and convective thinning of continental litho- Kent, D.V., and Irving, E., 2010, Influence of inclination
Cordillera and Pacific Margin: Geological Survey of
sphere: Geophysical Journal International, v.  128, error in sedimentary rocks on the Triassic and Juras-
Canada Paper 93-1A, p. 235–244, doi:​10​.4095​/134210​.
p. 125–150, doi:​10​.1111​/j​.1365​-246X​.1997​.tb04075​.x​. sic apparent pole wander path for North America and
Manuszak, J.D., Ridgway, K.D., Trop, J.M., and Gehrels,
Humphreys, E.D., Schmandt, B., Bezada, M.J., and Perry- implications for Cordilleran tectonics: Journal of Geo-
G.E., 2007, Sedimentary record of the tectonic growth
Houts, J., 2015, Recent craton growth by slab stacking physical Research, ser. B, Solid Earth, v. 115, B10103,
doi:​10​.1029​/2009JB007205​. of a collisional continental margin: Upper Juras-
beneath Wyoming: Earth and Planetary Science Letters,
Kistler, R.W., and Peterman, Z.E., 1978, Reconstruction sic–Lower Cretaceous Nutzotin Mountains sequence,
v. 429, p. 170–180, doi:​10​.1016​/j​.epsl​.2015​.07​.066​.
Ingersoll, R.V., 2008, Subduction-related sedimentary ba- of Crustal Blocks of California on the Basis of Initial eastern Alaska Range, Alaska, in Ridgway, K.D., Trop,
sins of the USA Cordillera, in Miall, A.D., ed., Sedi- Strontium Isotopic Compositions of Mesozoic Gran- J.M., Glen, J.M.G., and O’Neill, J.M., eds., Tectonic
mentary Basins of the World, Volume  5: Amsterdam, ite Rocks: U.S. Geological Survey Professional Paper Growth of a Collisional Continental Margin: Crustal
Netherlands, Elsevier, p. 395–428, doi:​10​.1016​/S1874​ 1071, 17 p. Evolution of Southern Alaska: Geological Society of
-5997​(08)00011​-7​. Kleinspehn, K.L., 1985, Cretaceous sedimentation and tec- America Special Paper  431, p.  345–377, doi:​10​.1130​
Ingersoll, R.V., 2012, Composition of modern sand and tonics, Tyaughton-Methow Basin, southwestern British /2007​.2431​(14)​.
Cretaceous sandstone derived from the Sierra Nevada, Columbia: Canadian Journal of Earth Sciences, v. 22, Massey, N.W.D., 1986, Metchosin Igneous Complex, south-
California, USA, with implications for Cenozoic and p. 154–174, doi:​10​.1139​/e85​-014​. ern Vancouver Island: Ophiolite stratigraphy devel-
Mesozoic uplift and dissection: Sedimentary Geology, Knight, E., Schneider, D.A., and Ryan, J., 2013, Thermo- oped in an emergent island setting: Geology, v.  14,
v. 280, p. 195–207, doi:​10​.1016​/j​.sedgeo​.2012​.03​.022​. chronology of the Yukon-Tanana terrane, west-central p.  602–605, doi:​ 10​.1130​/0091​-7613​(1986)14​<602:​
Ingersoll, R.V., and Schweickert, R.A., 1986, A plate- Yukon: Evidence for Jurassic extension and exhuma- MICSVI>2​.0​.CO;2​.
tectonic model for Late Jurassic ophiolite genesis, tion in the northern Canadian Cordillera: The Journal McClelland, W.C., Gehrels, G.E., and Saleeby, J.B., 1992,
Nevadan orogeny and forearc initiation, northern of Geology, v. 121, p. 371–400, doi:​10​.1086​/670721​. Upper Jurassic–Lower Cretaceous basinal strata along
California: Tectonics, v.  5, p.  901–912, doi:​10​.1029​ Lambert, R.S.J., and Chamberlain, V.E., 1988, Cordilleria the Cordilleran margin: Implications for the accretion-
/TC005i006p00901​. revisited, with a three-dimensional model for Creta- ary history of the Alexander-Wrangellia-Peninsular
Ingersoll, R.V., Dickinson, W.R., and Graham, S.A., 2003, ceous tectonics in British Columbia: The Journal of terrane: Tectonics, v.  11, p.  823–835, doi:​ 10​.1029​
Remnant-ocean submarine fans: Largest sedimentary Geology, v. 96, p. 47–60, doi:​10​.1086​/629192​. /92TC00241​.
systems on Earth, in Chan, M.A., and Archer, A.W., Laskowski, A.K., DeCelles, P.G., and Gehrels, G.E., 2013, McCrory, P.A., and Wilson, D.S., 2013, A kinematic model
eds., Extreme Depositional Environments: Mega End Detrital zircon geochronology of Cordilleran retroarc for the formation of the Siletz-Crescent forearc terrane
Members in Geologic Time: Geological Society of foreland basin strata, western North America: Tec­ by capture of coherent fragments of the Farallon and
America Special Paper 370, p.  191–208, doi:​10​.1130​ tonics, v. 32, p. 1027–1048, doi:​10​.1002​/tect​.20065​. Resurrection plates: Tectonics, v. 32, p. 718–736, doi:​
/0​-8137​-2370​-1​.191​. Lawton, T.F., and Garza, R.S.M., 2014, U-Pb geochronology 10​.1002​/tect​.20045​.
Israel, S., Murphy, D., Bennett, V., Mortensen, J., and Crow- of the type Nazas Formation and superjacent strata, Mihalynuk, M.G., and Friedman, R.M., 2006, U-Pb iso­
ley, J., 2010, New insights into the geology and mineral northeastern Durango, Mexico: Implications of a Ju- topic ages of intrusive-related mineralization north of
potential of the Coast belt in southwestern Yukon, in rassic age for continental-arc magmatism in north-cen- Terrace, BC, in Geological Fieldwork 2005: British
MacFarlane, K.E., Weston, L.H., and Relf, C., eds., tral Mexico: Geological Society of America Bulletin, Columbia Ministry of Energy, Mines and Petroleum
Yukon Exploration and Geology: Yukon Geological v. 126, p. 1181–1199, doi:​10​.1130​/B30827​.1​. Resources Paper 2006-1, p. 109–116.
Survey, p. 101–123. Li, C., van der Hilst, R.D., Engdahl, E.R., and Burdick, S., Mihalynuk, M.G., Smith, M.T., Hancock, K.D., and Dudka,
Jakob, J., and Johnston, S., 2015, New insights into the North 2008, A new global model for P wave speed variations S., 1994, Regional and economic geology of the Tulse-
American Cordillera forearc: Cretaceous to Eocene in Earth’s mantle: Geochemistry Geophysics Geosys- quah River and glacier areas (104K/ 12 & 13), in Geo-
tectonic evolution of the Leech River Schist, south- tems, v. 9, Q05018, doi:​10​.1029​/2007GC001806​. logical Fieldwork 1993: British Columbia, Ministry of
ern Vancouver Island, Canada: European Geosciences Li, S., Chung, S.-L., Wilde, S.A., Wang, T., Xiao, W.-J., Energy, Mines and Petroleum Resources Paper 1994-1,
Union General Assembly Conference Abstracts, v. 17, and Guo, Q.-Q., 2016, Linking magmatism with col- p. 171–197.

1518 Geological Society of America Bulletin, v. 129, no. 11/12

 Mantle and geological evidence for a Late Jurassic–Cretaceous suture spanning North America

Mihalynuk, M.G., Johnston, S.T., English, J.M., Cordey, America Special Paper 299, p.  159–171, doi:​10​.1130​ Ribe, N.M., Stutzmann, E., Ren, Y., and van der Hilst, R.,
F., Villeneuve, M.J., Rui, L., and Orchard, M.J., 2003, /SPE299​-p159​. 2007, Buckling instabilities of subducted lithosphere
­Atlin TGI Part II: Regional geology and mineralization Newberry, R.J., Layer, P.W., Burleigh, R.E., and Solie, D.N., beneath the transition zone: Earth and Planetary Sci-
of the Nakina area (NTS 104N/2W and 3), in Geologi- 1998, New 40Ar/39Ar dates for intrusions and mineral ence Letters, v.  254, p.  173–179, doi:​10​.1016​/j​.epsl​
cal Fieldwork 2002: British Columbia Ministry of En- prospects in the eastern Yukon-Tanana terrane, Alaska; .2006​.11​.028​.
ergy and Mines Paper 2003-1, p. 9–37. regional patterns and significance, in Gray, J.E., and Riddell, J.M., 1991, Geology of the Mesozoic volcanic and
Mihalynuk, M.G., Erdmer, P., Ghent, E.D., Cordey, F., Archi­ Riehle, J.R., eds., Geologic Studies in Alaska by the sedimentary rocks east of Pemberton, British Colum-
bald, D.A., Friedman, R.M., and Johannson, G.G., U.S. Geological Survey, 1996: U.S. Geological Survey bia, in Current Research, Part A, p. 245–254.
2004, Coherent French Range blueschist: Subduction Professional Paper 1595, p. 131–159. Ridgway, K.D., Trop, J.M., Nokleberg, W.J., Davidson,
to exhumation in <2.5  m.y.?: Geological Society of Nixon, G.T., and Orr, A.J., 2007, Recent revisions to the C.M., and Eastham, K.R., 2002, Mesozoic and Ceno-
America Bulletin, v.  116, p.  910–922, doi:​10​.1130​ early Mesozoic stratigraphy of Vancouver Island and zoic tectonics of the eastern and central Alaska Range:
/B25393​.1​. metallogenic implications, in Geological Fieldwork Progressive basin development and deformation in
Molnar, P., and Houseman, G.A., 2013, Rayleigh-Taylor 2006: British Columbia Ministry of Energy Mines and a suture zone: Geological Society of America Bul-
instability, lithospheric dynamics, surface topography Petroleum Resources Paper 2007-1, p. 163–177. letin, v.  114, p.  1480–1504, doi:​10​.1130​/0016​-7606​
at convergent mountain belts, and gravity anomalies: Nixon, G.T., Archibald, D.A., and Heaman, L.M., 1993, (2002)114​<1480:​MACTOT>2​.0​.CO;2​.
Journal of Geophysical Research–Solid Earth, v. 118, 40
Ar-39Ar and U-Pb geochronometry of the Polaris Riggs, N.R., Reynolds, S.J., Lindner, P.J., Howell, E.R.,
p. 2544–2557, doi:​10​.1002​/jgrb​.50203​. Alaskan-type complex, British Columbia; precise Barth, A.P., Parker, W.G., and Walker, J.D., 2013,
Monger, J.W.H., 2014, Seeking the suture: The Coast-Cas- timing of Quesnellia–North America interaction, in The early Mesozoic Cordilleran arc and Late Triassic
cade conundrum: Geoscience Canada, v.  41, p.  379– Geological Association of Canada–Mineralogical As- paleotopography: The detrital record in Upper Trias-
398, doi:​10​.12789​/geocanj​.2014​.41​.058​. sociation of Canada–Canadian Geophysical Union sic sedimentary successions on and off the Colorado
Monger, J.W.H., and Price, R.A., 1979, Geodynamic evo- Joint Annual Meeting, Waterloo, Ontario, Canada: Plateau: Geosphere, v.  9, p.  602–613, doi:​ 10​
.1130​
lution of the Canadian Cordillera; progress and prob- Geological Association of Canada Program with Ab- /GES00860​.1​.
lems: Canadian Journal of Earth Sciences, v.  16, stracts, v. 1993, p. 76. Rioux, M., Hacker, B., Mattinson, J., Kelemen, P., Blusz-
p. 770–791, doi:​10​.1139​/e79​-069​. Nolet, G., and Dahlen, F.A., 2000, Wave front healing and tajn, J., and Gehrels, G., 2007, Magmatic development
Monger, J.W.H., and Ross, C.A., 1971, Distribution of fusu­ the evolution of seismic delay times: Journal of Geo- of an intra-oceanic arc: High-precision U-Pb zircon
linaceans in the western Canadian Cordillera: Cana- physical Research–Solid Earth, v.  105, p.  19,043– and whole-rock isotopic analyses from the accreted
dian Journal of Earth Sciences, v. 8, p. 259–278, doi:​ 19,054, doi:​10​.1029​/2000JB900161​. ­Talkeetna arc, south-central Alaska: Geological Soci-
10​.1139​/e71​-026​. Obayashi, M., Yoshimitsu, J., Nolet, G., Fukao, Y., Shiobara, ety of America Bulletin, v. 119, p. 1168–1184, doi:​10​
Monger, J.W.H., Souther, J.G., and Gabrielse, H., 1972, H., Sugioka, H., Miyamachi, H., and Gao, Y., 2013, .1130​/B25964​.1​.
Evolution of the Canadian Cordillera; a plate-tectonic Finite frequency whole mantle P wave tomography: Ritsema, J., Deuss, A., Van Heijst, H.J., and Woodhouse,
model: American Journal of Science, v.  272, p.  577– Improvement of subducted slab images: Geophysical J.H., 2011, S40RTS: A degree-40 shear-velocity model
602, doi:​10​.2475​/ajs​.272​.7​.577​. Research Letters, v.  40, p.  5652–5657, doi:​10​.1002​ for the mantle from new Rayleigh wave dispersion,
Monger, J.W.H., Price, R.A., and Tempelman-Kluit, D.J., /2013GL057401​. teleseismic traveltime and normal-mode splitting func-
1982, Tectonic accretion and the origin of the two O’Neill, C., Mueller, D., and Steinberger, B., 2005, On the tion measurements: Geophysical Journal International,
­major metamorphic and plutonic welts in the Canadian uncertainties in hot spot reconstructions and the sig- v.  184, p.  1223–1236, doi:​10​.1111​/j​.1365​-246X​.2010​
.04884​.x​.
Cordillera: Geology, v. 10, p. 70–75, doi:​10​.1130​/0091​ nificance of moving hot spot reference frames: Geo-
Roeske, S.M., Mattinson, J.M., and Armstrong, R.L., 1989,
-7613​(1982)10​<70:​TAATOO>2​.0​.CO;2​. chemistry Geophysics Geosystems, v. 6, Q04003, doi:​
Isotopic ages of glaucophane schists on the Kodiak
Monger, J.W.H., Price, R.A., and Tempelman-Kluit, D.J., 10​.1029​/2004GC000784​.
Islands, southern Alaska, and their implications for
1983, Comment and Reply on “Tectonic accretion and Orchard, M., Struik, L.C., Rui, L., Bamber, E.W., Mamet,
the Mesozoic tectonic history of the Border Ranges
the origin of the two major metamorphic and plutonic B.L., Sano, H., and Taylor, H., 2001, Palaeontological
fault system: Geological Society of America Bul-
welts in the Canadian Cordillera”: Reply: Geology, and biogeographical constraints on the Carbonifer-
letin, v.  101, p.  1021–1037, doi:​10​.1130​/0016​-7606​
v.  11, p.  428–429, doi:​10​.1130​/0091​-7613​(1983)11​ ous to Jurassic Cache Creek terrane in central British
(1989)101​<1021:​IAOGSO>2​.3​.CO;2​.
<428:​CAROTA>2​.0​.CO;2​. Columbia: Canadian Journal of Earth Sciences, v. 38,
Rubin, C.M., Saleeby, J.B., Cowan, D.S., Brandon, M.T.,
Monger, J.W.H., Van der Heyden, P., Journeay, J.M., p. 551–578, doi:​10​.1139​/e00​-120​.
and McGroder, M.F., 1990, Regionally extensive mid-
Evenchick, C.A., and Mahoney, J.B., 1994, Jurassic– Pană, D.I., and van der Pluijm, B.A., 2015, Orogenic pulses
Cretaceous west-vergent thrust system in the north-
Cretaceous basins along the Canadian Coast belt: Their in the Alberta Rocky Mountains: Radiometric dating of western Cordillera; implications for continent-margin
bearing on pre–mid-Cretaceous sinistral displace- major faults and comparison with the regional tectono- tectonism: Geology, v.  18, p.  276–280, doi:​10​.1130​
ments: Geology, v. 22, p. 175–178, doi:​10​.1130​/0091​ stratigraphic record: Geological Society of America /0091​-7613​(1990)018​<0276:​REMCWV>2​.3​.CO;2​.
-7613​(1994)022​<0175:​JCBATC>2​.3​.CO;2​. Bulletin, v. 127, p. 480–502, doi:​10​.1130​/B31069​.1​. Sager, W.W., Handschumacher, D.W., Hilde, T.W.C., and
Montelli, R., Nolet, G., Dahlen, F.A., and Masters, G., Paterson, I.A., and Harakal, J.E., 1974, Potassium-argon Bracey, D.R., 1988, Tectonic evolution of the northern
2006, A catalogue of deep mantle plumes: New re- dating of blueschists from Pinchi Lake, central British Pacific plate and Pacific-Farallon-Izanagi triple junc-
sults from finite-frequency tomography: Geochemistry Columbia: Canadian Journal of Earth, v. 11, p. 1007– tion in the Late Jurassic and Early Cretaceous (M21–
Geophysics Geosystems, v.  7, Q11007, doi:​10​.1029​ 1011, doi:​10​.1139​/e74​-097​. M10): Tectonophysics, v.  155, p.  345–364, doi:​ 10​
/2006GC001248​. Pavlis, G.L., Sigloch, K., Burdick, S., Fouch, M.J., and .1016​/0040​-1951​(88)90274​-0​.
Moores, E.M., 1970, Ultramafics and orogeny, with models Vernon, F.L., 2012, Unraveling the geometry of the Saleeby, J.B., 1981, Ocean floor accretion and volcanoplu-
of the US Cordillera and the Tethys: Nature, v.  228, Farallon plate: Synthesis of three-dimensional imaging tonic arc evolution of the Mesozoic Sierra Nevada,
p. 837–842, doi:​10​.1038​/228837a0​. results from USArray: Tectonophysics, v.  532–535, in Ernst, W.G., ed., The Geotectonic Development of
Moores, E.M., 1998, Ophiolites, the Sierra Nevada, p. 82–102, doi:​10​.1016​/j​.tecto​.2012​.02​.008​. California: Rubey Volume 1: Englewood Cliffs, New
“Cordilleria,” and orogeny along the Pacific and Phillips, W.M., Walsh, T.J., and Hagen, R.A., 1989, Eocene Jersey, Prentice-Hall, Inc., p. 132–181.
Caribbean margins of North and South America: Inter- Transition from Oceanic to Arc Volcanism, Southwest Saleeby, J.B., 1983, Accretionary tectonics of the North
national Geology Review, v. 40, p. 40–54, doi:​10​.1080​ Washington: U.S. Geological Survey Open-File Re- American Cordillera: Annual Review of Earth and
/00206819809465197​. port 89-178, p. 89–178. Planetary Sciences, v.  11, p.  45–73, doi:​ 10​
.1146​
Morgan, W.J., 1981, Hotspot tracks and the opening of the Poulton, T.P., Hall, R.L., and Callomon, J.H., 1994, Ammo- /annurev​.ea​.11​.050183​.000401​.
Atlantic and Indian Oceans, in Emiliani, C., ed., The nite and bivalve assemblages in Bathonian through Ox- Saleeby, J.B., and Dunne, G., 2015, Temporal and tectonic
Oceanic Lithosphere: New York, John Wiley and Sons, fordian strata of northern Bowser Basin, northwestern relations of early Mesozoic arc magmatism, southern
v. 7, p. 443–487. British Columbia, Canada: Geobios, v. 27, p. 415–421, Sierra Nevada, California, in Anderson, T.H., Didenko,
Mortensen, J.K., and Jilson, G.A., 1985, Evolution of the doi:​10​.1016​/S0016​-6995​(94)80162​-2​. A.N., Johnson, C.L., Khanchuk, A.I., and MacDonald,
Yukon-Tanana terrane: Evidence from southeastern Ray, G.E., 1986, The Hozameen fault system and related J.H., Jr., eds., Late Jurassic Margin of Laurasia—A
Yukon Territory: Geology, v.  13, p.  806–810, doi:​10​ Coquihalla serpentine belt of southwestern British Record of Faulting Accommodating Plate Rotation:
.1130​/0091​-7613​(1985)13​<806:​EOTYTE>2​.0​.CO;2​. Columbia: Canadian Journal of Earth Sciences, v. 23, Geological Society of America Special Paper 513,
Müller, R.D., Sdrolias, M., Gaina, C., and Roest, W.R., 2008, p. 1022–1041, doi:​10​.1139​/e86​-103​. p. 223–268, doi:​10​.1130​/2015​.2513​(05)​.
Age, spreading rates, and spreading asymmetry of the Ren, Y., Stutzmann, E., van der Hilst, R.D., and Besse, J., Saleeby, J.B., Geary, E.E., Paterson, S.R., and Tobisch,
world’s ocean crust: Geochemistry Geophysics Geo- 2007, Understanding seismic heterogeneities in the O.T., 1989, Isotopic systematics of Pb/U (zircon) and
systems, v. 9, Q04006, doi:​10​.1029​/2007GC001743​. lower mantle beneath the Americas from seismic tomog­ 40
Ar/39Ar (biotite-hornblende) from rocks of the central
Murphy, D.C., van der Heyden, P., Parrish, R.R., Klepacki, raphy and plate tectonic history: Journal of Geophysi- Foothills terrane, Sierra Nevada, California: Geologi-
D.W., McMillan, W., Struik, L.C., and Gabites, J., cal Research–Solid Earth, v. 112, B01302, doi:​10​.1029​ cal Society of America Bulletin, v. 101, p. 1481–1492,
1995, New geochronological constraints on Jurassic /2005JB004154​. doi:​10​.1130​/0016​-7606​(1989)101​<1481:​ISOPUZ>2​
deformation of the western edge of North America, Ribe, N.M., 2010, Bending mechanics and mode selection .3​.CO;2​.
southeastern Canadian Cordillera, in Miller, D.M., and in free subduction: A thin-sheet analysis: Geophysical Schaeffer, A.J., and Lebedev, S., 2015, Global heterogene-
Busby, C., eds., Jurassic Magmatism and Tectonics of Journal International, v. 180, p. 559–576, doi:​10​.1111​ ity of the lithosphere and underlying mantle: A seis-
the North American Cordillera: Geological Society of /j​.1365​-246X​.2009​.04460​.x​. mological appraisal based on multimode surface-wave

Geological Society of America Bulletin, v. 129, no. 11/12 1519

 Sigloch and Mihalynuk

dispersion analysis, shear-velocity tomography, and Earth and Environmental Science Transactions of the van de Zedde, D.M.A., and Wortel, M.J.R., 2001, Shallow
tectonic regionalization, in Khan, A., and Deschamps, Royal Society of Edinburgh, v. 79, p. 105–121. slab detachment as a transient source of heat at mid-
F., eds., The Earth’s Heterogeneous Mantle: Switzer- Silver, E.A., and Smith, R.B., 1983, Comparison of terrane lithospheric depths: Tectonics, v. 20, p. 868–882, doi:​
land, Springer International Publishing, p.  3–46, doi:​ accretion in modern Southeast Asia and the Meso­ 10​.1029​/2001TC900018​.
10​.1007​/978​-3​-319​-15627​-9_1​. zoic North American Cordillera: Geology, v.  11, van der Heyden, P.A.H., 1989, U-Pb and K-Ar Geochro-
Schellart, W.P., 2008, Kinematics and flow patterns in deep p.  198–202, doi:​10​.1130​/0091​-7613​(1983)11​<198:​ nometry of the Coast Plutonic Complex, 53°N to 54°N,
mantle and upper mantle subduction models: Influence COTAIM>2​.0​.CO;2​. British Columbia, and Implications for the Insular-Inter-
of the mantle depth and slab to mantle viscosity ratio: Sisson, V., and Onstott, T., 1986, Dating blueschist meta- montane Superterrane Boundary [Ph.D. thesis]: Vancou-
Geochemistry, Geophysics, Geosystems, v. 9, Q03014, morphism—A combined Ar-40/Ar-39 and electron- ver, Canada, University of British Columbia, 515 p.
doi:​10​.1029​/2007GC001656​. microprobe approach: Geochimica et Cosmochimica van der Heyden, P.A.H., 1992, A Middle Jurassic to Early
Schmandt, B., and Humphreys, E., 2010, Complex subduc- Acta, v.  50, p.  2111–2117, doi:​10​.1016​/0016​-7037​ Tertiary Andean-Sierran arc model for the Coast belt
tion and small-scale convection revealed by body-wave (86)90264​-4​. of British Columbia: Tectonics, v. 11, p. 82–97, doi:​10​
tomography of the western United States upper mantle: Smith, P.L., 2006, Paleobiogeography and Early Jurassic .1029​/91TC02183​.
Earth and Planetary Science Letters, v.  297, p.  435– molluscs in the context of terrane displacement in van der Hilst, R.D., Widiyantoro, S., and Engdahl, E.R.,
445, doi:​10​.1016​/j​.epsl​.2010​.06​.047​. western Canada: Paleogeography of the North Ameri- 1997, Evidence for deep mantle circulation from
Schmandt, B., and Humphreys, E., 2011, Seismically im- can Cordillera, in Haggart, J., Enkin, R.J., and Monger, global tomography: Nature, v.  386, p.  578–584, doi:​
aged relict slab from the 55 Ma Siletzia accretion to the J.W.H., eds., Paleogeography of the North American 10​.1038​/386578a0​.
northwest United States: Geology, v.  39, p.  175–178, Cordillera: Evidence For and Against Large-Scale Dis- van der Lee, S., and Nolet, G., 1997, Seismic image of the
doi:​10​.1130​/G31558​.1​. placements: Geological Association of Canada Special subducted trailing fragments of the Farallon plate: Na-
Schweickert, R.A., 2015, Jurassic evolution of the western Paper 46, p. 81–94. ture, v. 386, p. 266–269, doi:​10​.1038​/386266a0​.
Sierra Nevada metamorphic province, in Anderson, Snow, C.A., and Scherer, H., 2006, Terranes of the western van der Meer, D.G., Spakman, W., van Hinsbergen, D.J.J.,
T.H., Didenko, A.N., Johnson, C.L., Khanchuk, A.I., Sierra Nevada Foothills metamorphic belt, California: Amaru, M.L., and Torsvik, T.H., 2010, Towards abso-
and MacDonald, J.H., Jr., eds., Late Jurassic Margin A critical review: International Geology Review, v. 48, lute plate motions constrained by lower-mantle slab
of Laurasia—A Record of Faulting Accommodating p. 46–62, doi:​10​.2747​/0020​-6814​.48​.1​.46​. remnants: Nature Geoscience, v.  3, p.  36–40, doi:​10​
Plate Rotation: Geological Society of America Spe- Snow, C.A., Wakabayashi, J., Ernst, W.G., and Wooden, J.L., .1038​/ngeo708​.
cial Paper 513, p.  299–358, doi:​10​.1130​/2015​.2513​ 2010, Detrital zircon evidence for progressive under- van der Meer, D.G., Torsvik, T.H., Spakman, W., van Hins-
(08)​. thrusting in Franciscan metagraywackes, west-central bergen, D.J.J., and Amaru, M.L., 2012, Intra-Pan­
Schweickert, R.A., and Cowan, D.S., 1975, Early Mesozoic California: Geological Society of America Bulletin, thalassa Ocean subduction zones revealed by fossil
tectonic evolution of the western Sierra Nevada, Cali- v. 122, p. 282–291, doi:​10​.1130​/B26399​.1​. arcs and mantle structure: Nature Geoscience, v.  5,
fornia: Geological Society of America Bulletin, v. 86, Staples, R.D., Gibson, H.D., Berman, R.G., Ryan, J.J., and p. 215–219, doi:​10​.1038​/ngeo1401​.
p. 1329–1336, doi:​10​.1130​/0016​-7606​(1975)86​<1329:​ Colpron, M., 2013, A window into the Early to mid- Van der Voo, R., Spakman, W., and Bijwaard, H., 1999a,
EMTEOT>2​.0​.CO;2​. Cretaceous infrastructure of the Yukon-Tanana terrane Mesozoic subducted slabs under Siberia: Nature,
Schweickert, R.A., Bogen, N.L., Girty, G.H., Hanson, R.E. recorded in multi-stage garnet of west-central Yukon, v. 397, p. 246–249, doi:​10​.1038​/16686​.
and Merguerian, C., 1984, Timing and structural ex- Canada: Journal of Metamorphic Geology, v.  31, Van der Voo, R., Spakman, W., and Bijwaard, H., 1999b,
pression of the Nevadan orogeny, Sierra Nevada, Cali- p. 729–753, doi:​10​.1111​/jmg​.12042​. Tethyan subducted slabs under India: Earth and Plan-
fornia: Geological Society of America Bulletin, v. 95, Stegman, D.R., Farrington, R., Capitanio, F.A., and Schel- etary Science Letters, v.  171, p.  7–20, doi:​10​.1016​
p.  967–979, doi:​10​.1130​/0016​-7606​(1984)95​<967:​ lart, W.P., 2010, A regime diagram for subduction /S0012​-821X​(99)00131​-4​.
TASEOT>2​.0​.CO;2​. styles from 3-D numerical models of free subduction: Villeneuve, M., Whalen, J.B., Anderson, R.G., and Struik,
Seton, M., Müller, R.D., Zahirovic, S., Gaina, C., Torsvik, Tectonophysics, v.  483, p.  29–45, doi:​10​.1016​/j​.tecto​ L.C., 2001, The Endako batholith: Episodic plutonism
T., Shephard, G., Talsma, A., Gurnis, M., Turner, M., .2009​.08​.041​. culminating in formation of the Endako porphyry
Maus, S., and Chandler, M., 2012, Global continental Steinberger, B., and Torsvik, T.H., 2008, Absolute plate mo- molybdenite deposit, north-central British Columbia:
and ocean basin reconstructions since 200 Ma: Earth- tions and true polar wander in the absence of hotspot Economic Geology and the Bulletin of the Society of
Science Reviews, v.  113, p.  212–270, doi:​10​.1016​/j​ tracks: Nature, v.  452, p.  620–623, doi:​ 10​
.1038​ Economic Geologists, v. 96, p. 171–196.
.earscirev​.2012​.03​.002​. /nature06824​. Whalen, J.B., Anderson, R.G., Struik, L.C., and Villeneuve,
Shephard, G.E., Mueller, R.D., and Seton, M., 2013, The Steinberger, B., Torsvik, T.H., and Becker, T.W., 2012, Sub- M.E., 2001, Geochemistry and Nd isotopes of the
tectonic evolution of the Arctic since Pangea breakup: duction to the lower mantle—A comparison between François Lake plutonic suite, Endako batholith: Host
Integrating constraints from surface geology and geo- geodynamic and tomographic models: Solid Earth, and progenitor to the Endako molybdenum camp, cen-
physics with mantle structure: Earth-Science Reviews, v. 3, p. 415–432, doi:​10​.5194​/se​-3​-415​-2012​. tral British Columbia: Canadian Journal of Earth Sci-
v.  124, p.  148–183, doi:​10​.1016​/j​.earscirev​.2013​.05​ Stewart, J.H., Anderson, T.H., Haxel, G.B., Silver, L.T., and ences, v. 38, p. 603–618, doi:​10​.1139​/e00​-080​.
.012​. Wright, J.E., 1986, Late Triassic paleogeography of the Wheeler, J.O., and McFeely, P., 1991, Tectonic Assemblage
Sigloch, K., 2008, Multiple-Frequency Body-Wave Tomog- southern Cordillera: The problem of a source for volu- Map of the Canadian Cordillera and Adjacent Parts
raphy [Ph.D. thesis]: Princeton, New Jersey, Princeton minous volcanic detritus in the Chinle Formation of the of the United States of America: Geological Survey of
University, 249 p. Colorado Plateau region: Geology, v. 14, p. 567–570, Canada Map 1712A, scale 1: 2,000,000.
Sigloch, K., 2011, Mantle provinces under North America doi:​10​.1130​/0091​-7613​(1986)14​<567:​LTPOTS>2​.0​ Woods, M.T., and Davies, G.F., 1982, Late Cretaceous gen-
from multifrequency P-wave tomography: Geochem- .CO;2​. esis of the Kula plate: Earth and Planetary Science
istry Geophysics Geosystems, v. 12, Q02W08, doi:​10​ Surpless, K.D., Sickmann, Z.T., and Koplitz, T.A., 2014, Letters, v. 58, p. 161–166, doi:​10​.1016​/0012​-821X​
.1029​/2010GC003421​. East-derived strata in the Methow Basin record rapid (82)90191​-1​.
Sigloch, K., and Mihalynuk, M.G., 2013, Intra-oceanic mid-Cretaceous uplift of the southern Coast Mountains Wu, J., Suppe, J., Lu, R., and Kanda, R., 2016, Philip-
subduction shaped the assembly of Cordilleran North batholith: Canadian Journal of Earth Sciences, v.  51, pine Sea and East Asian plate tectonics since 52 Ma
America: Nature, v.  496, p.  50–56, doi:​1 0​.1038​ p. 339–357, doi:​10​.1139​/cjes​-2013​-0144​. constrained by new subducted slab reconstruction
/nature12019​. Tobisch, O.T., Paterson, S.R., Saleeby, J.B., and Geary, methods: Journal of Geophysical Research–Solid
Sigloch, K., McQuarrie, N., and Nolet, G., 2008, Two-stage E.E., 1989, Nature and timing of deformation in the Earth, v.  121, no.  6, p.  4670–4741, doi:​1 0​.1002​
subduction history under North America inferred from Foothills terrane, central Sierra Nevada, California: Its /2016JB012923​.
multiple-frequency tomography: Nature Geoscience, bearing on orogenesis: Geological Society of America
v. 1, p. 458–462, doi:​10​.1038​/ngeo231​. Bulletin, v.  101, p.  401–413, doi:​10​.1130​/0016​-7606​ Science Editor: Bradley S. Singer
Silberling, N.J., Jones, D.L., Monger, J.W.H., and Coney, (1989)101​<0401:​NATODI>2​.3​.CO;2​. Associate Editor: Clinton P. Conrad
P.J., 1992, Lithotectonic Terrane Map of the North Umhoefer, P.J., Schiarizza, P., and Robinson, M., 2002,
American Cordillera: U.S. Geological Survey IMAP Relay Mountain Group, Tyaughton–Methow Basin, Manuscript Received 9 March 2016
2176, 2 sheets. southwest British Columbia: A major Middle Jurassic Revised Manuscript Received 4 March 2017
Silver, L.T., and Chappell, B.W., 1988, The Peninsular to Early Cretaceous terrane overlap assemblage: Cana- Manuscript Accepted 20 April 2017
Ranges Batholith: an insight into the evolution of the dian Journal of Earth Sciences, v.  39, p.  1143–1167,
Cordilleran batholiths of southwestern North America: doi:​10​.1139​/e02​-031​. Printed in the USA

1520 Geological Society of America Bulletin, v. 129, no. 11/12