- Email: [email protected]
Early central American forearc follows the subduction initiation rule
Journal Pre-proof Early Central American forearc follows the subduction initiation rule Scott A. Whattam, Camilo Montes, Robert J. Stern PII:
S1342-937X(19)30283-7
DOI:
https://doi.org/10.1016/j.gr.2019.10.002
Reference:
GR 2231
To appear in:
Gondwana Research
Received Date: 12 October 2018 Revised Date:
11 September 2019
Accepted Date: 2 October 2019
Please cite this article as: Whattam, S.A., Montes, C., Stern, R.J., Early Central American forearc follows the subduction initiation rule, Gondwana Research, https://doi.org/10.1016/j.gr.2019.10.002. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.
Comparison with SIRO lavas
Comparison with FAB
BON
BON
ARC
600
V Ti/
=2
Izu-Bonin Mariana FAB Exp. 352 FAB Exp. 351 FA-like basalts
Ti/ V
SIRO LU basalt SIRO UU bas-andesite
Ti/ V
800
=1
=1
0
0
1000
ARC
0
MORB
400
= Ti/V
V Ti/
=2
0
MORB
50
= Ti/V
WPB
200
50
WPB
a
d
Golfito OP (Ha) (75-66 Ma)
Grp I Grp II (enr)
Chagres-Bayano OP
BON
SIRO UU
600
ARC
SIRO LU
V Ti/
=2
0
MORB
400
= Ti/V
50
WPB
200
e
b =1
0
1000 Sona-Azuero proto-arc & arc (73-39 Ma) others Grp III
800
Ti/ V
V ppm
Sona-Azuero this study
Grp I Grp II (enriched)
Ti/ V
Sona-Azuero OP (89-85 Ma) others
800
=1
0
1000
Sona-Azuero this study Grp III
600
ARC V Ti/
BON
Golfito proto-arc (75-66 Ma, B)
=
20
MORB
Chagres-Bayano arc (70-39 Ma)
400
= Ti/V
50
WPB
200
Group I (from b)
0
5
10
15
f
c
0 20
25 0
5
10
Ti ppm/1000
Graphical Abstract Whattam et al. 2019
15
20
25
1
Early Central American forearc follows the subduction initiation rule
2 3 4 5
Scott A. Whattama, Camilo Montesb, and Robert J. Sternc
6 7 8 9 10 11 12 13 14 15 16 17
a
Department of Geosciences, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia; Department of Physics and Geology, Universidad del Norte, Barranquilla, Colombia c Department of Geosciences, University of Texas at Dallas, Richardson, TX, USA b
Words: 7752 excluding abstract, acknowledgements and appendices
CONTACT Scott A. Whattam [email protected]
18
Keywords: Central American forearc; subduction initiation; subduction initiation rule; forearc;
19
volcanic arc; ophiolite
1
20
Abstract
21
The “subduction initiation rule“ (SIR) (Whattam and Stern, 2011) advocates that proto-arc and
22
forearc complexes preserved in ophiolites and forearcs follow a predictable chemotemporal
23
and/or chemostratigraphic vertical progression. This chemotemporal evolution is defined by a
24
progression from bottom to top, from less to more depleted and slab-metasomatized sources.
25
This progression has been recently documented for other igneous suites associated with
26
subduction initiation. The Sona-Azuero forearc complex of southern Panama represents the
27
earliest magmatic arc activity at the Central American Volcanic Arc system. Comparison of new
28
and existing geochemical data for the circa 82-40 Ma Sona-Azuero Proto-Arc/Arc, its underlying
29
89-85 Ma “oceanic plateau” of SW Panama and the 72-69 Ma Golfito Proto-Arc of southern
30
Costa Rica with the 70-39 Ma Chagres-Bayano Arc of eastern Panama exhibits a chemotemporal
31
progression as described above and which follows the SIR. Sona-Azuero lavas are
32
predominantly MORB-like, whereas those of the younger Chagres-Bayano complex are mostly
33
VAB-like; lavas of the Golfito Proto-Arc typically show characteristics intermediate to that of
34
the Sona-Azuero and Chagres-Bayano proto-arc/arc complexes. On the basis of isotope evidence
35
as shown in other studies, lava types of all three complexes are clearly derived from a source
36
contaminated by the Caribbean Large Igneous Province plume; we term these “plume-
37
contaminated” forearc basalts and volcanic arc basalts, respectively. Apart from a plume-induced
38
subduction initiation origin for the Panamanian forearc, these insights suggest otherwise similar
39
petrogenetic origins and tectonic setting to lavas comprising earliest-formed forearc crust, and
40
most ophiolites, which follow the SIR.
41 42
1. Introduction
2
43 44
Ophiolites are ancient, on-land vestiges of oceanic lithosphere ascribed to seafloor spreading to
45
form a nascent proto-forearc in the earliest stages of subduction initiation. As shown by Whattam
46
and Stern (2011), Tethyan-type supra-subuduction zone ophiolites (Pearce et al., 1984; Pearce,
47
2003) of the Eastern- Mediterranean-Persian Gulf region (Robertson, 2004) exhibit similar
48
chemostratigraphic/chemotemporal progressions (changes in lava chemistry with time) as that of
49
forearc lithosphere of classic records of magmatism during subduction initiation, such as the Izu-
50
Bonin-Mariana intra-oceanic arc system (Reagan et al., 2010; Ishizuka et al., 2011). Tethyan-
51
type ophiolites for example, in Oman (Semail), Greece (Pindos) and Albania (Mirdita) formed
52
by initial eruption of MOR-like tholeiitic basalts (i.e., forearc basalts, FAB, Reagan et al., 2010)
53
followed by calc-alkaline volcanic arc basalt (VAB) or basaltic-andesite lavas and intrusives
54
characterized by large ion lithophile-enrichment and high-field strength element-depletion,
55
which were sometimes followed by latest-stage boninites (e.g., Alabaster et al., 1982; Saccani
56
and Photiades, 2004; Dilek et al., 2008). This exact sequence is not universal but a common
57
theme is emerging from global examples. Variations in subduction initiation progression is
58
revealed by the Troodos ophiolite which lacks MORB-like tholeiites and instead shows a
59
progression from earlier-formed basalt-andesite-dacite-rhyolite series backarc basin-like basalts
60
to boninites (Pearce and Robinson, 2010). At the Izu-Bonin-Mariana forearc, generation of
61
earliest MORB-like FAB during initial subduction initiation was followed by eruption of
62
boninites (Reagan et al., 2010; 2017); from this time forward igneous activity retreated to the site
63
of the long-term magmatic front. A similar chemotemporal progression has recently been
64
documented for the nascent stages of Greater Antilles Arc system construction (Torró et al.,
65
2017). This chemotemporal progression reflects complementary magmatic and mantle responses
3
66
(MORB VAB affinities and fertile lherzolite depleted harzburgite, respectively) during
67
subduction initiation and encapsulates the ‘subduction initiation rule’ (SIR) of Whattam and
68
Stern (2011). The SIR stipulates that the composition of subduction initiation lavas change from
69
less to more high-field strength depleted and from less to more slab-metasomatized with time,
70
i.e., lavas change from MORB-like to VAB-like over the course of subduction initiation, and this
71
rule links forearcs formed during subduction initiation and ophiolites. This chemotemporal
72
progression evolution reflects formation of proto-forearc/forearc lithosphere as a result of mantle
73
melting that is increasingly influenced by enrichments from the sinking slab during subduction
74
initiation. Mid-oceanic ridge-like FAB lavas are the consequence of decompression melting of
75
upwelling asthenosphere and mark the initial magmatic manifestation of subduction initiation.
76
Contribution of fluids from dehydrating oceanic crust and sediments on the sinking slab is minor
77
during the early stages of subduction initiation, but continuous melting results in a depleted,
78
harzburgitic residue. This residue is increasingly metasomatized by fluids from the sinking slab
79
and later partial melting of this residue yields ‘typical’ suprasubduction-zone-like lavas with
80
high-field strength element depletions and large ion element enrichments in the latter stages of
81
subduction initiation. Recognizing the SIR has profound implications for the development of
82
robust tectonic plate reconstructions and interpreting and understanding forearc formation during
83
subduction initiation. Identification of the SIR in ophiolite complexes implies formation similar
84
to that of forearc lithosphere via seafloor spreading during subduction initiation (Stern and
85
Bloomer, 1992; Stern et al., 2012).
86
The purpose of this paper is not to provide a regional treatment of the tectonic
87
evolution of or new stratigraphic constraints on the early Central American forearc but rather, to
88
examine the early Central American forearc as an extension to cases for which the SIR has been 4
89
tested so far. We do this by focusing on Late Cretaceous and Early Paleogene igneous rocks of
90
Central America and compare their chemotemporal (changes in chemistry with time) progression
91
with the chemotemporal progression of other complexes associated with subduction initiation
92
(forearcs and ophiolites).
93
Crucial to our chemotemporal arguments is recognition that the Sona-Azuero and Golfito proto-
94
arc/arc segments in central Panama and SW Costa Rica, respectively, began to form at least 5
95
M.y. prior to establishment of the Chagres-Bayano Arc in central Panama (see Section 2 for
96
Regional setting and the Supplementary Document for description of these complexes). We
97
document the availability of absolute and relative ages which provides the basis of this
98
recognition in the Supplementary Document (Table S1, see also Table 1).
99 100
2. Regional setting
101
As a result of detailed multidisciplinary studies of Late Cretaceous ‘oceanic igneous
102
terranes’ (terminology of Hauff et al., 1997) and arc-related complexes exposed in the forearc of
103
the Central American Volcanic Arc ( Fig. 1a) in Costa Rica (e.g., Buchs et al., 2010) and Panama
104
(Fig. 1b) (e.g., Hauff et al., 2000a; Tomascak et al., 2000; Geldmacher et al., 2001, 2008; Lissina
105
, 2005; Gazel et al., 2009, 2015; Wörner et al., 2009; Buchs et al., 2009, 2010, 2011; Wegner et
106
al., 2011; Montes et al., 2012a, 2012b; Whattam et al., 2015a, b) we can re-evaluate the
107
chemotemporal evolution of these oldest Central American Volcanic forearc complexes in the
108
context of subduction initiation processes. Most researchers agree that the bulk of the Caribbean
109
Plate is composed of the Caribbean Large Igneous Province (CLIP) (Fig. 1a), an expansive
110
oceanic “plateau” that formed over the Galapagos hotspot prior to its northeasterly drift and
111
insertion between North and South America (see Kerr et al., 2000) prior to subduction initiation.
5
112
Relatedly, the CLIP is interpreted by most as underlying younger arc segments not only in Costa
113
Rica and Panama but also along NW South America in western Ecuador and Colombia all the
114
way to the Leeward Antilles in Curaçao and Aruba (see references in Whattam and Stern,
115
2015a). The ‘standard’ model holds that volcanic arc construction occurred upon the trailing
116
edge of the CLIP in present-day Costa Rica and Panama (Pindell and Kennan, 2009; Wegner et
117
al., 2009; Buchs et al., 2010; Wegner et al., 2011) and elsewhere in the Caribbean realm along
118
the southern margin of the Caribbean Plate in NW South America and the Leeward Antilles. In
119
contrast, on the basis of collated geochemical, isotopic and radioisotopic age datasets of these
120
‘plume-and-arc-related’ complexes in Central America and NW South America which exhibit
121
compositional and temporal overlap between Late Cretaceous hotspot-derived CLIP magmatism
122
and the magmatic effects of subduction initiation, Whattam and Stern (2015a) suggested arc
123
construction in a single, rapidly evolving hybrid plume- and subduction-related environment. By
124
‘plume- and subduction- related’ we mean proto-forearc/forearc complexes that were generated
125
synchronously or nearly so as a result of interaction between the plumehead and the (soon to be)
126
subduction-modified lithosphere that it interacted with; as a result, the complexes which
127
constitute this study were generated from a mantle source that was plume-dominated at first and
128
evolved with time to be modified by subduction. According to this model, arc formation
129
occurred subsequent to subduction initiation which was catalyzed by plume emplacement
130
(plume-induced subduction initiation). A plume-induced subduction initiation mechanism has
131
recently been shown to be valid on the basis of numerical thermomechanical modelling (Gerya et
132
al., 2015).
133
Although the Farallon Plate has been subducting beneath the western Americas since at
134
least 200 Ma, subduction initiation at the Central American Arc was independent of Farallon
6
135
Plate subduction. Central America subduction initiation began between 90-75 Ma along the
136
western periphery of the Caribbean Large Igneous Province (CLIP), which was emplaced within
137
the middle of the Farallon Plate beginning about 92 Ma (Sinton et al., 1997; Kerr et al., 2003 and
138
references therein). Some previous studies of the Central American forearc igneous complexes
139
discussed in this study [see Supplementary Document] suggest that subduction initiation began
140
~75-71 Ma with the magmatic arc growing on pre-existing oceanic plateau (CLIP) basement
141
(Buchs et al. 2010; Wegner et al. 2011]. In contrast, Lissinna et al. (2006) suggest that the infant
142
Panamanian island arc on the western side of the CLIP dates back to 89 Ma. The studies of
143
Lissinna et al. (2006) and Wegner et al. (2011) did not posit a cause for subduction initiation but
144
Buchs et al. (2010) suggested this was tectonically induced, caused by compression of the
145
thickened Caribbean Plate during westward migration of the Americas and collision with South
146
America. As explained above, Whattam and Stern (2015a) in contrast suggested that subduction
147
initiation was the result of plume emplacement and discuss why an induced subduction initiation
148
scenario was unlikely (see Section 5.4; see also Section 5.5 for a discussion of different
149
subduction initiation mechanisms). We note that the timeframe of the Whattam and Stern
150
(2015a) model, which posits subduction initiation as the result of plume emplacement at 90 Ma,
151
is consistent with the conclusions of Lissinna et al. (2006) who suggest subduction initiation by
152
89 Ma.
153
The active Central American volcanic front stretches ~1100 km along the western margin
154
of the Caribbean plate from Costa Rica through Nicaragua, El Salvador and Guatemala to the
155
Guatemala-Mexico border at the southern margin of the North American plate (Figure 1a). The
156
locus of magmatic activity has shifted from east to west after the Eocene. Early plume- and
157
subduction related sequences in the east comprise the ~89-40 Ma Sona-Azuero (Azuero
7
158
Marginal Complex of Buchs et al. 2010), 72-69 Ma Golfito, and 70-39 Ma Chagres-Bayano
159
complexes (Figure 1b, see also Table 1, Table S1 and the Supplementary Document for
160
description of these complexes). The Sona-Azuero and Chagres-Bayano complexes are
161
interpreted as being floored by the 89-85 Ma CLIP (Wegner et al., 2009; Buchs et al., 2010;
162
Wegner et al., 2011; Montes et al., 2012) and the Golfito Complex has been interpreted as both
163
CLIP oceanic plateau (Hauff et al., 2000) and more recently as arc (Buchs et al., 2010). Prior to
164
the study of Buchs et al. (2010), it was generally assumed that most oceanic forearc units along
165
western Costa Rica and Panama represented uplifted segments of the CLIP. Similar
166
interpretations have been made for many oceanic complexes along NW South America and the
167
Leeward Antilles (see Whattam et al., 2015a).
168 169
3. Methods
170
3.1. Petrography
171 172
Fifty new representative thin sections of the Sona-Azuero complex lavas were examined under a
173
petrographic microscope. Sample locations and coordinates are provided in Figure 2 and Table
174
S2, respectively, and selected micrographs comprise Supplementary Figure S1.
175 176
3.2. Elemental analysis
177
Whole-rock samples (n = 25) of the Azuero complex were crushed and ground to powders for
178
major, trace and rare earth element determinations (Table 2). Details of analytical procedures and
179
elemental analysis are provided in the Supplementary Document.
180
8
181
3.3 Data compilation and manipulation
182
3.3.1. Data compilation
183
For comparison with our 25 new whole-rock geochemical analyses of Azuero lavas (Table 2),
184
we also compile geochemical data of lavas of the Sona-Azuero (Buchs et al. 2010; Wegner et al.
185
2011), Golfito (Hauff et al., 2000; Buchs et al., 2010) and Chagres-Bayano complexes (Wegner
186
et al. 2011; Montes et al. 2012b). For further comparison with these Central America forearc
187
complex datasets, we also compile data of global MORB (Gale et al., 2013) and volcanic arc
188
basalts (VAB) of depleted arc systems (PetDB, https://www.earhchem.org/petdb and GEOROC,
189
(http://georoc.mpch-mainz.gwdg.de/georoc/; see Hawkesworth et al., 1997, for definition of
190
‘depleted arc systems’); Late Cretaceous to Eocene ‘plume and arc related’ units (Whattam and
191
Stern, 2015a) in Panama and Costa Rica; lavas from complexes around the circum-Caribbean
192
region interpreted as CLIP; SIR ophiolite lavas (Whattam and Stern, 2011); Izu-Bonin forearc
193
basalts (Reagan et al., 2010; Shervais et al., 2019; and Izu-Bonin ‘forearc-like’ basalts (Arculus
194
et al., 2015)).. The MORB dataset (Gale et al., 2013) is mostly basalt (mean of 50.4 wt. % SiO2)
195
but ranges from 44.9-70.4 wt. % SiO2. Volcanic and plutonic samples comprising Late
196
Cretaceous to Eocene (~85-39 Ma) ‘plume and arc-related’ units (Whattam and Stern 2015a) in
197
Panama and southernmost Costa Rica are considered as either oceanic plateau or proto-arc and
198
arc constructed upon the CLIP (e.g., Buchs et al., 2010) or hybrid plume- and arc-related
199
(Whattam and Stern, 2015a). Basalts interpreted as CLIP (n = 192 tholeiitic and 7 enriched
200
oceanic-island-like basalts) are compiled from GEOROC and are from the Caribbean Sea; the
201
western margin of the Caribbean Plate in Costa Rica (Nicoya, Herradura, Tortugal); the western
202
margin of NW South America in western Ecuador (Piñón Formation and the Pallatanga and
203
Pedernales-Esmeraldas unit) and western Colombia (Gorgona Island, Serrania de Baudo
9
204
Formation and the Western Cordillera); the Lesser Antilles in Aruba (Aruba Lava Formation);
205
and the Caribbean Sea (Colombian and Venezuelan Basins).
206 207
3.3.2. Data manipulation
208
In order to assure that suitable quality analyses of least altered samples of Sona-Azuero, Golfito
209
and Chagres-Bayano were used in our study, only samples which yield oxide totals of 96-102 wt.
210
% (excluding loss on ignition, LOI) are plotted. In our geochemical plots using major elements,
211
oxides are recalculated on an anhydrous (volatile-free) basis and normalized to 100%. However,
212
our geochemical plots and arguments rely primarily on trace elements known to be non-mobile
213
up to greenschist-facies metamorphic conditions.
214 215
4. Results
216
Our lavas are subdivided into three chemical groups on the basis of chondrite-normalized La/Yb
217
concentrations (see section 4.2.1 and Figure 5). Groups I, II and III are analogous to those
218
interpreted by others as plateau basalts, (volumetrically subordinate) enriched basalts and proto-
219
arc or arc basalts (Buchs et al., 2010; Wegner et al., 2011), respectively. As shown by Wegner et
220
al. (2011) and Whattam and Stern (2015a,b), Group I are tholeiitic, Group III range from
221
tholeiitic to calc-alkaline and Group II are analogous to oceanic island tholeiites (Supplementary
222
Figure 2).
223 224
4.1. Petrography
225
Most Group I MORB-like (see Section 5) Azuero lavas are fine-grained tholeiitic basalts.
226
Textures are typically intergranular to intersertal or subophitic and mineralogy is primarily
10
227
plagioclase, clinopyroxene and alteration products (Supplementary Figures S1a, b). In other very
228
fine-grained varieties, clinopyroxene is not obvious and the mineralogy is dominated by thin,
229
wispy, highly altered plagioclase laths. Other samples are coarser-grained and some of these,
230
with large, sericitized plagioclase laths and fresh, subhedral, interstitial clinopyroxene, are
231
texturally similar to Group III arc-like (see Section 5) basalts. The crystallization sequence of
232
plagioclase followed by clinopyroxene is typical of MORB (Bryan, 1983; Pearce et al., 1984).
233
Group II oceanic island basalt-like (see Section 5, Supplementary Figures S1c, d) lavas
234
are subordinate, making up ~10-20 % of the total Azuero samples in our dataset and the
235
combined datasets of Buchs et al. (2010) and Wegner et al. (2011). Overall, these basalts are
236
moderately to heavily altered and distinguished by their chocolate brown coloration and
237
ubiquitous calcite- and chlorite-filled amygdales in the coarser-grained, weakly porphyritic and
238
medium-grained plagioclase phyric samples. Fine-grained varieties of these basalts exhibit
239
micro-porphyritic texture defined by ~1 mm clinopyroxene and 2-5 mm plagioclase phenocrysts.
240
Apatite is an accessory phase (typically <5%).
241
Our Group III arc-like (see Section 5) Azuero lavas are rare (n = 3) and in contrast to
242
lavas of Groups I and II, usually display micro-porphyritic or unique to this group, robustly
243
porphyritic textures defined by large (1.5-2.0 mm), commonly euhedral, fresh clinopyroxene
244
phenocrysts (Supplementary Figures S1e, f). The crystallization sequence, in contrast to the
245
Group I MORB-like lavas, is clinopyroxene followed by plagioclase which is typical of hydrous
246
(i.e., arc) magmas in which early plagioclase crystallization is suppressed (Pearce et al. 1984;
247
Cameron 1985; Sisson and Grove 1994).
248 249
4.2. Major element chemistry
11
250
4.2.1. MgO vs. TiO2, Al2O3, FeOt and K2O
251
As developed below in Section 4.3, the Sona-Azuero lavas of Buchs et al. (2010), Wegner et al.
252
(2011) and our dataset are subdivided into three groups on the basis of chondrite-normalized
253
La/Yb concentrations. Group I are interpreted by Buchs et al. (2010) and Wegner et al. (2011) as
254
plateau basalts; Group II as enriched basalts (Buchs et al., 2010); and Group III as proto-arc or
255
arc volcanics (Buchs et al., 2010; Wegner et al., 2011). These groupings are referred to below.
256
A plot of SiO2 vs. MgO (Figure 3) shows that lavas considered in this study range from
257
basaltic to rhyolitic with MgO concentrations of 1.06 to 11.68 wt.%. Fractionation patterns for
258
MORB (n = 14, 782) and depleted (see Hawkesworth et al., 1997) arc systems (n = 4, 379) are
259
broadly similar but with the latter ranging to lower MgO at low SiO2 contents (˂55 wt.%) and
260
higher MgO at higher SiO2 (55-72 wt.%). Despite the considerable overlap of MORB and VAB,
261
a large number of Sona-Azuero Group III arc (n = 8) and Chagre-Bayano samples (n = 7) exhibit
262
higher arc-like MgO than MORB at silica concentrations ˃ 55 wt.% SiO2. As well, two Sona-
263
Azuero proto-arc, three Sona-Azuero Group II samples (including two from this study) and one
264
Chagres-Bayano sample display lower SiO2 than MORB for a given MgO concentration at low
265
SiO2 (45-48 wt. %) and fall within the arc domain.
266
In Figures 4a-d, we plot MgO vs. TiO2, Al2O3, FeOt and K2O for basalts of our Azuero
267
dataset alongside basalts of the Sona-Azuero, Golfito and Chagres-Bayano suites. For
268
comparison, also plotted on Figure 4 are lavas of MOR and depleted arc systems (basalts only).
269
In all plots there exists considerable overlap between MORB and VAB but some samples plot
270
uniquely within either field. In consideration of MgO vs. TiO2, all six Sona-Azuero Type II
271
basalts plot uniquely within or just above the MORB field and outside of the arc field. Two of
12
272
our three Type II Sona-Azuero basalts also have Al2O3 and FeOt concentrations which plot
273
within the MORB field. In MgO vs. TiO2 and the remainder of the MgO vs. major oxides plots
274
(Al2O3, FeOt and K2O), between two to four low-magnesium (~3-6 wt.% MgO) Chagre-Bayano
275
lavas plot uniquely within the arc field. Similarly, apart from K2O, one low MgO (~4.2 wt.%
276
Mg) Sona-Azuero proto-arc sample plots distinctively within the arc field; two of our Sona-
277
Azuero Type III arc lavas have TiO2 and FeOt akin to arc basalts. Additionally, two high MgO
278
(~10-12 wt. %) Chagre-Bayano lavas exhibit MgO vs. TiO2, Al2O3, FeOt and K2O compositions
279
that plot distinctly within the arc field or on the cusp between MORB and arc. The remainder of
280
samples plot within the overlapping area between MORB and VAB.
281
4.3. Trace element chemistry
282
4.3.1. MgO vs. Ce, Yb, Y and Zr
283
We also plot MgO vs. a light and heavy rare earth element (LREE, Ce and HREE, Yb) and two
284
high field strength elements (Y, Zr) for basalts of our Azuero dataset alongside basalts of the
285
Sona-Azuero, Golfito and Chagres-Bayano suites for comparison with MORB and depleted arc
286
system basalts (Figures 4e-h). In terms of MgO vs. Ce, two of our Sona-Azuero high-TiO2 Group
287
II samples plot above both the MORB and arc fields with the highest Ce of all samples (46-47
288
ppm) whereas two Sona-Azuero Group II samples of Buchs et al. (2009) plot within the MORB
289
field along with one Sona-Azuero proto-arc and arc sample. Many low-magnesium (˂6 wt.%
290
MgO) basalts plot within the arc field including two of our Sona-Azuero Group III arc lavas,
291
three Sona-Azuero Group I lavas interpreted as plateau by Buchs et al. (2009) and Wegner et al.
292
(2011), three Sona-Azuero Group III arc basalts and five Chagres-Bayano basalts. On the other
293
end of the MgO spectrum, one high-MgO (~11 wt. % MgO) Group I Sona-Azuero and one
13
294
Chagres-Bayano basalt (~12 wt. % MgO) also plot within the arc field. The remaining basalts
295
plot where the MORB and arc fields overlap in MgO vs. Ce space.
296
On a plot of MgO vs. Yb (Fig. 4f), four Sona-Azuero Group III arc basalts including all
297
three of our Sona-Azuero Group III basalts plot alongside six Chagres-Bayano basalts uniquely
298
within the low-HREE arc field. Also, three Sona-Azuero Group I, six Sona-Azuero Group III
299
proto-arc and arc and six Chagre-Bayano basalts plot outside of both the MORB and arc field;
300
the remaining majority of samples plot within the MORB realm with Yb ˃ ~1.4.
301
Many samples exhibit arc-like HFSE concentrations. On a MgO vs. Y plot (Fig. 4g), our
302
three Sona-Azuero Group III basalts, four of six Sona-Azuero Group II basalts and eight Sona-
303
Azuero Group III proto-arc and arc basalts fall alongside almost half of the Chagre-Bayano
304
basalts (12 of 26 samples) within the arc field. Our three Sona-Azuero Group III basalts and
305
Chagres-Bayano basalts exhibit the lowest Y concentrations (˂14 ppm). The remaining basalts
306
fall within an area of MORB and arc overlap. A similar trend is seen on MgO vs. Zr (Fig. 4h)
307
with our Sona-Azuero Group III and Chagres-Bayano basalts exhibiting the lowest Zr arc-like
308
contents of all samples.
309
4.3.2. Subdivision of Sona-Azuero lavas into three groups on the basis of [La/Yb] CN
310
Table 3 provides the foundation for subdivision of our Azuero lavas into three groups on the
311
basis of chondrite-normalized La/Yb concentrations ([La/Yb]CN).
312
Compositionally, our Azuero Group I basalts are slightly to moderately light rare earth
313
element depleted ([La/Yb]CN = 0.75-1.09, mean = 0.96) and compositionally analogous to the
314
Group I (non-enriched) Azuero ‘plateau’ samples of Buchs et al. (2010) and the ‘CLIP’ basalts
315
of Wegner et al. (2011) ([La/Yb]CN = 0.63-1.15, mean = 0.92, Table 3).
14
316
Our Group II Azuero basalts are volumetrically subordinate, light rare earth element-
317
enriched ([La/Yb]CN = 2.29-2.86) and compositionally similar to the Group II ‘enriched plateau’
318
samples of Buchs et al. (2010) ([La/Yb]CN = 1.92-2.81) (Table 3). Taking all enriched samples
319
into consideration (i.e., of Buchs et al., 2010 and ours), these basalts are perhaps best described
320
as oceanic island tholeiites (OIT on Supplementary Figure 2) on the basis of TiO2 vs. Y/Nb
321
relations (Floyd and Winchester, 1975; Winchester and Floyd, 1977).
322
On the basis of chondrite-normalized rare earth element and N-MORB-normalized plots
323
(see below), our Group III lavas are VAB-like and analogous to the Azuero Proto-Arc and Arc
324
samples of Buchs et al. (2010) and the ‘CLIP’ Arc samples of Wegner et al. (2011), although our
325
samples range to lower absolute rare earth element concentrations. Rare earth element patterns of
326
Group III lavas range from slightly light rare earth element-enriched ([La/Yb]CN = 1.28-1.58],
327
Table 3) to flat to slightly ‘U-shaped’ (see next section), i.e., middle rare element-depleted, e.g.,
328
two samples have [La/Gd]CN > 1 and [Yb/Gd]CN > 1).
329 330
4.3.3. Chondrite-normalized rare earth element and N-MORB-normalized incompatible element
331
plots
332 333
Chondrite-normalized rare earth element and N-MORB-normalized plots of Sona-Azuero,
334
Golfito and Chagres-Bayano basalts are provided in Figs. 5 and 6, respectively.
335
Although the rare earth element patterns of Golfito and Chagres-Bayano basalts are
336
similar to those of Sona-Azuero there are some significant differences. Rare earth element
337
patterns of proto-arc lavas of the Golfito Complex (Fig. 5d) generally fall within the range of the
338
Group I and Group III Sona-Azuero lavas interpreted as oceanic plateau and proto-arc/arc
15
339
respectively, but range to very low absolute rare earth element abundances (mean ∑ rare earth
340
element of 32) and light rare earth element fractionations (e.g., [La/Yb]CN of 0.81). These values
341
are higher only than that of the Chagres-Bayano Arc (mean ∑ rare earth element ~30) and the
342
lone Chagres-Bayano basalt interpreted as oceanic plateau ([La/Yb]CN of 0.47) (Figs. 5e, f).
343
Relatedly, this [La/Yb]CN ratio of the lone Chagres-Bayano basalt interpreted as oceanic plateau
344
(0.47) is much lower than that of the mean of Sona-Azuero oceanic plateau and arc basalts (with
345
mean [La/Yb]CN of 0.93 and 1.54, respectively). Although the Chagres-Bayano Arc exhibits
346
mean [La/Yb]CN intermediate to that of the Sona-Azuero oceanic plateau and proto-arc/arc (mean
347
[La/Yb]CN of 1.18 vs. 0.93 and 1.54, respectively), the absolute rare earth element abundances of
348
Chagres-Bayano Arc lavas range to much lower concentrations (mean ∑ rare earth elements of
349
30) of ~ 3x chondrite (Fig. 5f).
350
In terms of N-MORB-normalized incompatible element patterns, the most conspicuous
351
differences between our Azuero Groups are: the lack of large ion lithophile element enrichment
352
in fluid-mobile elements (e.g, Rb, Ba, K, Pb) in Group II, which is well developed in both
353
Groups I and III; the lack of a negative Nb-anomaly (with respect to Th and Ce) in Groups I and
354
II, but which is obvious in Group III; and the pattern of Y and the heavy rare earth elements (Yb,
355
Lu), which are flat in Groups I and III but which decrease with decreasing incompatibility in
356
Group II (Figs. 6a-c). The differences between the N-MORB-normalized patterns of the Sona-
357
Azuero basalts interpreted as CLIP plus our Group I basalts interpreted as segments of the
358
earliest hybrid, plume-dominated unit and basalts interpreted as arc are modest, apart from the
359
prominent Nb depletion and the slightly more, fluid-modified compositions of the latter. For
360
example, although the CLIP basalts exhibit no obvious negative Nb-depletion, they do exhibit
361
modest negative Zr and Ti anomalies and contain only slightly less Ba and Pb (Fig. 6a). 16
362
N-MORB normalized plots of Golfito and Chagres-Bayano basalts are provided in Figs.
363
6d-f. Golfito proto-arc basalts fall completely within the range of Sona-Azuero oceanic plateau
364
and proto-arc/arc basalts but range to lower concentrations of large ion lithophile elements (Ba,
365
K, Pb and Sr) (Fig. 6d). Although a negative Nb anomaly is not well developed in the lone
366
Golfito sample interpreted as oceanic plateau (Hauff et al., 2000) it does exhibit a subtle negative
367
Zr anomaly and concentrations of U, Pb, K and Sr higher than or similar to Golfito basalts
368
interpreted as arc (Buchs et al., 2010); the Golfito proto-arc samples of Buchs et al. (2010)
369
exhibit subtle but clear negative Nb anomalies (Fig. 6d)). The N-MORB-normalized
370
incompatible element plot of the lone Chagres-Bayano sample interpreted as oceanic plateau
371
(Fig. 6e) is somewhat anomalous as it lacks the high LILE enrichments exhibited by most other
372
arc (e.g., Chagres-Bayano Arc, Fig. 6f) and oceanic plateau samples.
373 374
5. Discussion
375
In the following sections, we explore the significance of our new geochemical data and that of
376
other workers to address the following questions: (1) Do Late Cretaceous – Paleogene igneous
377
rocks of Panama and Costa Rica follow the “subduction initiation rule” promulgated for the Izu-
378
Bonin-Mariana arc and many ophiolites?; (2) What is the nature of the contact between “oceanic
379
plateau” and arc and chemotemporal evolution of magmatism?; and (3) What was the nature of
380
the early arc system?
381 382
5.1 Evidence of adherence of the early Central American Volcanic Arc system to the
383
subduction initiation rule
17
384
We focus here on elements that are not mobilized by hydrothermal alteration and metamorphism
385
for the purpose of constraining mantle source fertility, degrees of partial melting and mantle
386
source fugacity. Ratios of Zr/Y, Nb/Y (and Nb/Yb) are useful as these decrease with mantle
387
depletion and as partial melting increases. Because of the compatibility of Y and Yb in garnet,
388
these ratios can also reflect if garnet was present during melt generation but this may be the case
389
only in Group II enriched oceanic island tholeiite-like basalts which exhibit low Y and heavy
390
rare earth element relative to middle rare earth element ( Fig. 5b). In Figures 7-9, in addition to
391
plotting Panamanian and Costa Rican samples from this study, we also plot SIR ophiolite basalts
392
and basaltic andesites for comparison. We do this because SIR ophiolite lavas are (i) thought to
393
have formed during SI in a proto-forearc environment and show a distinctive chemotemporal
394
(change in chemistry with time) trend from less- to more-depleted and subduction–modified
395
compositions with time (i.e., MORB- to VAB-like) (Whattam and Stern, 2011) which (ii) along
396
with other chemical criteria (see below), are critical for understanding the evolution of the
397
mantle sources for Sona-Azuero and Chagres-Bayano.
398 399
5.1.1. Source fertility
400
In Fig. 7, we plot Zr vs Zr/Y. As demonstrated by Whattam and Stern (2011) and shown in Fig.
401
7a, whereas the older lavas of the earliest-formed lower subduction initiation rule ophiolite units
402
are mostly MORB-like (84% of the lower unit basalts plot as MORB), those of the younger,
403
uppermost unit in contrast plot completely within the VAB field.
404
These relations show that forearc units progress from tapping/melting of a less to more
405
depleted source or lower to higher degrees of partial melting over the course of seafloor
406
spreading during subduction initiation. Notably, all tholeiiitic Group I Sona-Azuero basalts
18
407
interpreted as CLIP (Fig. 7b) plot within the compositional field defined by all (Group III) Sona-
408
Azuero and most Chagres-Bayano and Golfito basalts interpreted as proto-arc (Fig. 7c)
409
suggesting a similar source or degrees of partial melting for Sona-Azuero oceanic plateau and
410
arc. As well, Fig. 7c confirms what was shown by the N-MORB and rare earth element plots,
411
i.e., that Chagres-Bayano arc basalts were derived from a more depleted source, or one that has
412
undergone higher degrees of partial melting than Sona-Azuero. Similarly, the Golfito basalts
413
exhibit evidence of derivation from a source intermediate in composition to that of Sona-Azuero
414
and Chagres-Bayano and Sona-Azuero Group III basalts are more similar in composition to
415
Sona-Azuero Group I than to Chagres-Bayano. Geochemical relations demonstrate a MORB to
416
VAB transition from the Sona-Azuero to the Chagres-Bayano unit and demonstrate adherence to
417
the SIR for the early Central American Volcanic forearc.
418 419
5.1.2. Degrees of partial melting
420
Fig. 7 shows that the source(s) of the Chagres-Bayano basalts and our Group III Azuero basalts
421
were either more depleted or underwent higher degrees of partial melting than the Sona-Azuero
422
source. These plots however, don’t distinguish between the two possibilities, and if the latter is
423
accurate, provide estimates of the relative degrees of partial melting.
424
Cr vs. Y (Fig. 8) can be used to determine relative degrees of partial melting of basalts
425
(Murton 1989). Superimposed on Fig. 8 are sources 1, 2 and 3 (S1, S2, S3) which represent a
426
MORB source, a more depleted source after 20% MORB extraction, and a most depleted source
427
after ~12% extraction of S2, respectively. We caution that the F values discussed below are
428
model dependent and we are interested not in absolute degrees of partial melting but differences
429
in partial melting between lavas of Sona-Azuero and Chagres-Bayano. Also shown in Fig. 8a are
19
430
fields defined by MORB, VAB and boninite (Pearce, 1982). Inspection of Fig. 4b reveals that
431
whereas the lower basalts of subduction initation rule ophiolites are consistent with ~20-40%
432
partial melting of S1 or 20% partial melting of S1 and ~10% partial melting of S2, the upper unit
433
basaltic andesites of subduction initation rule ophiolites indicate higher degrees of partial melting
434
(~10-40% partial melting of S2). A similar relation is seen for Sona-Azuero vs. Chagres-Bayano
435
lavas (Figs. 8c, d). Most Sona-Azuero basalts interpreted as oceanic plateau plot within the
436
MORB field and are consistent with ~20-40% partial melting of S1 (Fig. 8c). Approximately half
437
of the Sona-Azuero basalts interpreted as arc and some Chagres-Bayano arc basalts also plot
438
within the MORB field consistent with ~ up to 40% partial melting of S1, or perhaps more
439
realistically~ 12% melting of S2 (Fig. 8d). The other half of the Sona-Azuero basalts interpreted
440
as arc along with the Golfito basalts record partial melting estimates of 10-20% of S2. Compared
441
to SI ophiolites, the Chagres-Bayano Arc basalts clearly exhibit evidence of derivation from the
442
highest degrees of partial melting. About 35% of the Chagres-Bayano arc basalts plot within the
443
boninite field and are consistent with derivation via up to about 16-17% partial melting of S3.
444
Although boninites require >52 wt.% SiO2, >8 wt.% MgO and <0.5 wt.% TiO2 according to the
445
IUGS (LeBas, 2000), some of these Chagres-Bayano basalts which plot as boninite contain >8
446
wt.% MgO and <0.5 wt.% TiO2 (e.g., samples P-010223-6, PAN-03-016, Wegner et al., 2011).
447
Such high degrees of partial melting reflect exceptionally hot and shallow forearc mantle beneath
448
young intra-oceanic subduction-related systems (e.g., Tatsumi and Eggins, 1995). Like the Zr vs
449
Zr/Y plot (Fig. 7) the Cr/Y plot (Fig. 8) demonstrates an unequivocal chemotemporal evolution
450
that follows the SIR.
451 452
5.1.3. Source oxygen fugacity
20
453
It has long been recognized that MORB are distinguishable from VAB on the basis of Ti/V due
454
to the more oxidized nature (lower Ti/V ratio) of arc environments vs. the more reduced nature
455
of MORB sources (Shervais, 1982). We plot SIR ophiolite lavas vs. the older and younger
456
segments of the Sona-Azuero, Chagres-Bayano and Golfito complexes in Ti/V space in Figs. 9a-
457
c to demonstrate that the features developed above in various incompatible and immobile
458
elements hold true when considering source fugacity; i.e., whereas the older Sona-Azuero suite
459
(basalts interpreted as oceanic plateau and proto-arc/arc) is consistent with derivation from a
460
MORB-like source, basalts from Chagres-Bayano on the other hand are clearly more consistent
461
with derivation from a more arc-like source.
462
In Figs. 9d-f, we also plot FAB from the Izu-Bonin Mariana Arc system (Reagan et al.,
463
2010), Exp. 352 forearc basalts from the outer Izu-Bonin Mariana Arc system (Shervais et al.,
464
2019), and Exp. 351 Site U1438 (Arculus et al., 2015) forearc basalt-like lavas (which presently
465
occupy the rear arc of the Izu-Bonin Mariana Arc system). Lowermost basalts of the Izu-Bonin
466
Mariana forearc were first interpreted as trapped Phillipine Sea Plate (deBari et al., 1999), a
467
similar interpretation of which was applied to the Tonga forearc (Meffre et al., 2012). However,
468
these first-formed lavas (forearc basalt) are MORB-like and instead mark the first magmatic
469
expression of subduction initiation (Reagan et al., 2010). As shown in Fig. 9e, the vast majority
470
of the older MORB-like Sona-Azuero and the lone Chagres-Bayano basalt (all considered as
471
oceanic plateau) and Golfito basalts plot within the fields defined by the forearc basalts.
472
Similarly, the vast majority of the younger Chagres-Bayano Arc basalts also plot within the field
473
defined by forearc basalts (Fig. 9f), whereby most (77%) of the older Sona-Azuero basalts
474
interpreted as arc plot outside the fields defined by Izu-Bonin Mariana (Ishizuka et al., 2006;
475
Reagan et al., 2010) and Exp. 352 forearc basalts (Expedition 352 Scientists, Preliminary Report
21
476
2015) within the MORB field. This again demonstrates a SIR chemotemporal evolution from
477
early MORB-like lavas to later arc-like lavas, this time on the basis of oxygen fugacity, in the
478
early Central American forearc.
479 480
5.2. Isotopic constraints on the plume-contaminated nature of the source
481
A myriad of studies show the plume-contaminated nature of Central American forearc lavas as
482
the result of emplacement of the CLIP which immediately preceded and instigated subduction
483
initiation (plume-induced subduction initiation, Whattam and Stern (2015a) and references
484
therein). In terms of source components and source mixing, it was shown by Allibon et al.
485
(2008) that the isotopic compositions of Ecuadorian arc or arc-like lava sequences (which
486
according to Whattam and Stern 2015a are essentially analogous to arc lavas in Sona-Azuero,
487
Golfito and Chagres-Bayano) built upon CLIP oceanic plateau fragments could be explained in
488
terms of tripolar mixing of three components: a dominant depleted MORB mantle component; a
489
high µ (where µ is a 238U/204Pb ratio of an Earth reservoir) component carried by the CLIP; and a
490
subducted pelagic sediment component. It was further shown by Allibon et al. (2008) that an
491
average contribution of 70-75% depleted MORB mantle, 12-15% high µ , and 7-15% pelagic
492
sediments (cherts) is required to explain their composition. Furthermore, Whattam ( 2018)
493
showed that the Central American lavas in this study also plot within the
494
206
495
Cretaceous Ecuadorian lavas in the study of Allibon et al. (2008). It is for these reasons that we
496
use the term ‘plume contaminated’ to describe the forearc basalts of Sona-Azuero and the VAB
497
of Chagres-Bayano.
143
Nd/144Nd vs
Pb/204Pb field defined by the Galapagos Plume at 90 Ma similarly to the western Late
498
22
499
5.3. Nature of the contact between “oceanic plateau” and arc and chemotemporal evolution of
500
magmatism
501 502
Buchs et al. (2010) present a synthetic ‘tectonostratigraphic’ chart (their Fig. 3, reproduced in
503
part in our Fig. 10a) but there is a ~10-12 M.y. hiatus between inferred termination of oceanic
504
plateau construction at 85 Ma and incipient arc construction at 75-73 Ma (timing of SI as
505
interpreted by Buchs et al., 2010) at both the Azuero marginal and Golfito complexes. The only
506
study we know of that documents the nature of the contact between the lower “oceanic plateau
507
unit” and the upper arc unit in the Sona-Azuero Complex is that of Corral et al. (2011) (their Fig.
508
2, reproduced in our Fig. 10b). They (Corral et al. 2011) show a conformable contact between
509
the oceanic plateau basement and the overlying proto-arc sequence, which in turn is conformably
510
overlain by the arc sequence (the Rio Quema Formation, see Fig. 10b). This bolsters the
511
supposition of Whattam and Stern (2015a) of a continuous, uninterrupted magmatic succession
512
between underlying lavas interpreted as plateau and overlying ones interpreted as proto-arc/arc.
513
Lissinna et al. (2006) report that earliest arc activity commenced at 88.3±5 Ma but unfortunately,
514
no further details are available. If earliest formation of the Panamanian Arc was indeed ~90-89
515
Ma, this supports the model of Whattam and Stern (2015a) that plateau and arc magmatism
516
overlapped beginning about 90 Ma. We also note that Buchs et al. (2010) remark that some
517
earliest supra-subduction zone proto- igneous rocks are indistinguishable from those of ‘typical’
518
oceanic plateaus which also hints that plateau and arc magmatism overlapped (or that earliest SI
519
lavas tapped a strongly plume-contaminated source). Corral et al. (2011) provide a stratigraphic
520
section though the Rio Quema Formation (Azuero), showing that the Azuero proto-arc group
521
simply overlies the Azuero (oceanic plateau) igneous basement which, if the ages of Lissinna et
23
522
al. (2006) are accurate, suggest a continuous, uninterrupted interval from oceanic plateau to
523
proto-arc/arc magmatism. This, coupled with the complete compositional overlap between Sona-
524
Azuero lavas and intrusives interpreted as oceanic plateau and those interpreted as proto-arc and
525
arc suggest that both are related in space and time. Clearly, more stratigraphic and
526
geochronologic investigations are needed but the data in hand support a model whereby the
527
evolution from earlier CLIP-like basalts to younger arc-like lavas was continuous and
528
uninterrupted (Fig. 11a) as opposed to a scenario whereby arc construction occurred upon a pre-
529
existing plateau (Fig. 11b).
530
It is increasingly clear that the subdivision between oceanic plateau and arc magmatic
531
sequences in Central America are gradational. We recognize that the subdivision is based on
532
lavas and intrusives with negative primitive-mantle normalized high-field strength element
533
anomalies (particularly Nb) being assigned as arc and those without as oceanic plateau.
534
However, it was shown by Whattam and Stern (2011) that the first magmatic expression of some
535
subduction initation systems is MORB-like, and some basalts of lower unit subduction initiation
536
ophiolites exhibit weak primitive mantle normalized negative Nb-anomalies, others do not; a
537
robust negative Nb-anomaly and associated large ion lithophile enrichment is typically seen
538
only in younger lavas and intrusives. We agree with the interpretation of Whattam and Stern
539
(2015a) that the earliest magmatic activity at the Central American Arc system in Panama and
540
Costa Rica was the result of tapping of a hybrid, plume-subduction source as the result of plume-
541
induced subduction initiation. As shown in Sections 5.1.1.-5.1.3., whereas lavas of both Sona-
542
Azuero units - i.e., those interpreted as oceanic plateau and those interpreted as arc - are
543
‘MORB-like’, the lavas of the Chagres-Bayano unit are mostly ‘arc-like’. We conclude on this
24
544
basis that magmatism evolved continuously from plume-contaminated MORB- to VAB-like with
545
time consistent with the SIR.
546 547
5.4. Subduction initiation mechanisms
548
As explained by Stern (2004) and later by Stern and Gerya (2018), subduction initiation can be
549
either spontaneous or induced. In the case of the former, there are three possible types: transform
550
collapse, passive margin collapse, and plume head margin collapse (i.e., plume-induced
551
subduction initiation). In contrast, induced subduction initiation is triggered by continued plate
552
convergence following collision, for example of an oceanic plateau with a subduction zone.
553
There are two types of induced subduction initiation: a new subduction zone of similar dip to the
554
original subduction zone can initiate behind the collision (transference) or in front of the
555
collision and (polarity reversal).
556
The earliest workers to propose spontaneous subduction initiation at a lithospheric
557
weakness (transform separating older, denser lithosphere from younger, warmer lithosphere) was
558
Matsumoto and Tomoda (1983). This idea of spontaneous subduction initiation along a
559
transform fault was adopted by Stern and Bloomer (1992) to explain the early evolution of the
560
Izu Bonin Marianas arc system and by others for other regions of forearc formation (e.g.,
561
Whattam et al., 2006, 2008; Shafaii Moghadam et al., 2013; Zhou et al., 2018). Other studies
562
have focused on forced convergence across fracture zones (e.g., Hall et al., 2003). Subduction
563
nucleation along a passive margin (Marques et al., 2014; Ueda et al., 2008; Burov and Cloetingh,
564
2010; Van der Lee et al., 2008) is implied in the Wilson (1966) cycle but until very recently
565
(Pandey et al., 2019), there are no known Cenozoic examples (Stern, 2004). Pandey et al. (2019)
25
566
recovered forearc basalts and overlying boninitic-like forearc basalts in the Laxmi Basin, western
567
Indian passive margin, which record geochemical and isotopic attributes identical to first formed
568
lavas of the Izu Bonin Marianas and supra-subduction zone ophiolites and may provide the first
569
evidence for subduction initiation along a passive margin. Perhaps the best example of a induced
570
subduction initiation polarity reversal event is that caused by collision of the Ontong Java
571
Plateau, the world’s largest igneous province, which attempted to subduct to the SW at the Vitiaz
572
Trench ~10-4 Ma (Phinney et al., 1999; Cooper and Taylor, 1985). Thick, buoyant lithosphere of
573
the Plateau jammed the subduction zone and caused a subduction polarity flip to the NE. As
574
explained in the Introduction, Buchs et al. (2010) suggested that subduction initiation at the
575
Central American Arc was tectonically induced and resulted from compression of the thickened
576
Caribbean Plate during westward migration of the Americas and collision of the thickened
577
Caribbean Plate with South America. As explained above in Section 5.3, on the basis of a
578
comprehensive set of geochemical, isotopic and geochronologic data and tectonic considerations,
579
we concur with Whattam and Stern (2015) that arc construction in Central America began
580
instead about 90 Ma as the result of plume-induced subduction initiation. A second example of
581
plume-induced subduction initiation has been recently proposed for the Cascadia subduction
582
zone in the Pacific NW, USA (Stern and Dumitru, 2019) and plume-induced subduction
583
initiation has shown to be viable on the basis of numerical thermomechanical modelling (Gerya
584
et al., 2015).
585 586
5.5. Physical nature of the early arc system and tectonic considerations
587
In their study of Panamanian magmatism circa 75 Ma to present, Wegner et al. (2011) suggest
588
that arc magmatism began in the Sona and Azuero peninsulas at ~71 Ma (Sona-Azuero Arc), 26
589
then shifted to the Chagres-Bayano region at 66 Ma; Wegner et al. (2011) also apparently
590
consider these as two distinct arcs. We disagree with these notions for several reasons. Firstly,
591
arc lengths are typically not at the 100s of km scale but rather 1000s of km scale. For example,
592
the modern-day Central American Volcanic Arc system is ~1500 km long and the Izu-Bonin
593
Marianas Arc system is ~2800 km long. The magmatic arc ‘belts’ along the Sona-Azuero
594
peninsula and the Chagres-Bayano region are only ~200-250 km long, but restoration of
595
paleomagnetically derived vertical-axis rotations (Montes et al., 2012a) suggests that the
596
combined length of the Campanian to Eocene Central American arc was at least 1000 km.
597
Furthermore, combined geochemical and geochronological evidence suggests that the nascent
598
Central American arc stretched westwards from the Nicoya peninsula in western Costa Rica
599
some 200 km south of Nicaragua, to either present-day western Ecuador or Colombia some 1400
600
km to the east (Whattam and Stern, 2015a). Secondly, the age constraints of Lissinna (2005)
601
indicate that Sona-Azuero magmatic activity was contemporaneous with Chagres-Bayano
602
magmatism between 70-40 Ma. Thirdly, unequivocal ‘arc’ magmatism at the Sona-Azuero and
603
Golfito complexes began by at least 75 Ma (Buchs et al., 2010) but the 89-85 Ma Sona-Azuero
604
CLIP MORB-like oceanic plateau basalts are largely indistinguishable from the younger Sona-
605
Azuero arc and proto-arc basalts. Furthermore, Lissinna et al. (2006) suggest arc magmatism
606
began by ~90 Ma. Finally, as shown by Whattam and Stern (2015a), most (~90-85 Ma) oceanic
607
plateau basalts from the western edge of the CLIP (Costa Rica, e.g., Nicoya, Herradura,
608
Tortugal) including those of Sona-Azuero and the NW segment of the South American Plate in
609
Ecuador and Western Colombia, exhibit subduction modified compositions. Although a lateral
610
west to east transition in chemistry cannot be categorically ruled out, the relations in the
611
chemistry of Sona-Azuero vs. Chagres-Bayano basalts are more consistent with a
27
612
vertical/stratigraphic transition, although eruptions may have become more arc-like as
613
magmatism retreated from the trench. This chemostratigraphic progression can also be
614
considered as chemotemporal as the change in magmatic affinities is largely vertical and hence a
615
record also of change in chemistry with time; this chemotemporality is identical to that recorded
616
by the slightly older Sona-Azuero and the younger Chagres-Bayano complexes. Thus, we
617
suggest that the Chagres-Bayano arc segment represents a vertical as well as lateral migration of
618
arc magmatism that is also seen for Sona-Azuero circa 71 Ma (Wegner et al., 2011) (Figure 7).
619
In other words, the Chagres-Bayano arc segment may represent the carapace of the early
620
established Central American Arc system.
621
If Chagres-Bayano does indeed represent the latest phase of early Central American Arc
622
magmatism, the present position of the Chagres-Bayano arc segment indicates that it is offset to
623
the north. In the tectonic model of Whattam et al. (2012), the early (pre-40 Ma) Central
624
American Arc was shut down at about 40 Ma as the result of collision of an oceanic plateau with
625
the arc system (see also Kerr and Tarney, 2005). As explained by Montes et al. (2012) and
626
Whattam et al. (2012), this collision caused left-lateral offset of the originally contiguous early
627
(pre-40 Ma) Central American Volcanic Arc system which stretched from at least western Costa
628
Rica in the NW to western Ecuador or Colombia in the SE (Whattam and Stern, 2015a) and
629
resulted in a >100 km displacement to the north between 40-30 Ma (Lissinna, 2005; Montes et
630
al., 2012). Alternatively, left-lateral offset may have been caused by a shift in the subducted
631
plateau from Azuero to a region to the east (Lissinna et al., 2006) or as the result of collision of
632
South America with the Panamanian Arc system (Farris et al., 2011; Barat et al., 2014).
633
Regardless of the mechanism of displacement or the nature of the colliding body, we suggest that
28
634
the Chagres-Bayano arc segment was translated to the north after originally forming the carapace
635
and rear of the early arc system subsequent to slab rollback but prior to displacement (Fig. 11).
636 637
6. Conclusions
638 639
An unambiguous chemotemporal transition from early-formed plume-contaminated MORB-like
640
to VAB-like lavas is seen in Late Cretaceous-Paleogene magmatic arc sequences in Panama as
641
revealed by incompatible trace element compositions which illustrate fundamental differences
642
between the older Sona-Azuero complex and the younger Chagres-Bayano complex. While the
643
trace element signatures of basalts from both units are similar, the absolute REE concentrations
644
of Chagres-Bayano basalts range to much lower Σ rare earth element concentrations (12-52 ppm)
645
than those of Sona-Azuero (Σ rare earth elements = 33-102 ppm), exhibit more negative high
646
field strength element-anomalies and higher fluid mobile element/immobile element ratios.
647
Relative to Sona-Azuero, the 70-39 Ma Chagres-Bayano basalts reflect higher degrees of partial
648
melting of a more depleted and oxidized mantle source that received greater slab-derived shallow
649
and deep subduction contributions.
650
Moreover, the Sona-Azuero basalts interpreted as proto-arc and arc are compositionally
651
more similar to the Sona-Azuero basalts interpreted as oceanic plateau than they are to the
652
Chagres-Bayano basalts, which if age estimates are accurate suggest that CAVAS subduction
653
began by ~89 Ma. The overarching chemotemporal trend recorded by Sona-Azuero and Chagres-
654
Bayano sequences may reflect the combined effects of a west to east migration of magmatism at
655
68 Ma as well as a vertical/stratigraphic transition. We suggest that the arc segment now
656
preserved at Chagres-Bayano was originally generated above and to the rear of Sona-Azuero
29
657
forearc magmatism and was displaced to the north perhaps due to collision with an oceanic
658
plateau on the Cocos plate, which shut down the early arc system ~40 Ma.
659
The chemotemporal progression from older MORB-like lavas and intrusives at Sona-
660
Azuero to younger VAB-like lavas at Chagres-Bayano to the east confirms adherence of the Late
661
Cretaceous to Eocene Central American forearc to the SIR, similar to Tethyan ophiolites and the
662
Izu-Bonin-Mariana forearc. Importantly, we note that this adherence is independent of the two
663
competing tectonic models described. In other words, even if arc construction occurred intrusive
664
into and atop a pre-existing oceanic plateau and we exclude the MORB-like affinities of the
665
oceanic plateau, the SIR still applies as the Sona-Azuero Proto-Arc and arc lavas and intrusives
666
are also MORB-like in contrast to those of the Chagres-Bayano Arc which instead exhibit
667
unequivocal volcanic arc affinities.
668 669
Acknowledgements
670 671
We are grateful to Mark Tupper, NSF EAR-0824299, National Geographic, Smithsonian
672
Institution, and Ricardo Perez S.A. which partially funded this project. S. Zapata and D. Ramirez
673
are acknowledged for their help in the field and for generating the base map with the sample
674
localities. Access to field areas and collection permits were granted by Ministerio de Industria y
675
Comercio of Panamá. We thank Jörg Geldmacher, Jeffrey Ryan, and Yamirka Rojas-Agramonte
676
for their reviews of a previous version of the manuscript and Fernando Marques, Kosuke Ueda
677
and an anonymous reviewer for their reviews of the current manuscript and positive comments
678
which resulted in a greatly improved paper.
679
Figure Captions
30
680
Figure 1 (a) Present tectonic configuration of the Circum-Caribbean region. (b) Magnification
681
and detail of the boxed region in Figure 1a showing the distribution of Late Cretaceous to middle
682
Eocene (~75-39 Ma) arc magmatism in Golfito, easternmost Costa Rica, Sona-Azuero, southern
683
Panama and Chagres-Bayano, eastern Panama. Abbreviations in: (a) CR, Costa Rica,; GAA,
684
Greater Antilles Arc; LAA, Lesser Antilles Arc; PAN, Panama; PR, Puerto Rico. (b) CB,
685
Chagres-Bayano; CI, Coiba Island; Cord de Pan, Cordillera of Panama; GF, Golfito; PC, Panama
686
City; PCFZ, Panama Canal Fracture Zone.
687 688
Figure 2 Geologic map of the Azuero peninsula showing the location of our samples. Map is
689
modified from Dirección General de Recursos Minerales (1991), Buchs et al., (2010) and Corral
690
et al. (2011). See Figure 1 for broader location of the Azuero peninsula. Abbreviations: CB,
691
Chagres-Bayano; G, Group: GI, Group I; GII, Group II; GIII, Group III; Nic, Nicaragua; SA,
692
Sona-AzueroSee text for definitions of these groups.
693 694
Figure 3 MgO vs. SiO2 for all lavas of this study superimposed with the fields of MORB and
695
lavas of depleted arc systems (see text). Abbreivation: enr, enriched.
696 697
Figure 4 MgO vs. (a) TiO2, (b) Al2O3, (c) FeOt, (d) K2O, (e) Ce, (f) Yb, (g) Y and (h) Zr for all
698
lavas of this study superimposed with the fields of lavas of MORB and depleted arc systems
699
(basalts only)Abbreviation: enr, enriched.
700 701
Figure 5 Chondrite-normalized REE-normalized plots of Sona-Azuero lavas with 52 or less wt.
702
SiO2 from this study vs. Sona-Azuero, Golfito and Chagres-Bayano lavas with less than 52 wt. %
31
703
SiO2 from other studies. The similarity in chondrite-normalized LREE fractionations, i.e.,
704
[La/Yb]CN of our samples with those from other studies (Lissinna 2005; Buchs et al. 2010;
705
Wegner et al. 2011) in (a-c) form the basis for classification of our samples into three groups (see
706
Table 3 for further details): (i) Group I, LREE-depleted, (ii) Group II, LREE-enriched
707
(transitional to alkalic) and (iii) Group III, flat to slightly LREE-enriched. Chondrite
708
concentrations are from Nakamura (1974). Abbreviations: OP, oceanic plateau; PAR, plume- and
709
arc-related. The grey region in (b) to (f) represent the data presented in (a) and the dashed
710
regions in (d) to (f) represent the data presented in (c).
711 712
Figure 6 N-MORB-normalized plots of Sona-Azuero lavas with 52 or less wt. SiO2 from this
713
study vs. Sona-Azuero, Golfito and Chagres-Bayano lavas with less than 52 wt. % SiO2 from
714
other studies. N-MORB abundances are from Sun and McDonough (1989). Abbreviations: (a, e)
715
PAR, plume- and arc-related; (g, i) OP, oceanic plateau. The grey region in (b) to (f) represent
716
the data presented in (a) and the dashed regions in (d) to (f) represent the data presented in (c).
717 718
Figure 7 Zr vs. Zr/Y (Pearce and Norry 1979) relations of (a) subduction initiation rule ophiolite
719
(SIRO) basalts, (b) Sona-Azuero, Golfito and Chagres-Bayano basalts interpreted as oceanic
720
plateau and our Group I and II Azuero basalts and (c) Sona-Azuero, Golfito and Chagres-Bayano
721
basalts interpreted as arc and our Group III Azuero basalts. All Central American samples are
722
basalts only (i.e., no intrusives). Abbreviations in: (a) bas-and, basaltic andesite; DM, depleted
723
mantle; EM, enriched mantle; WPB, within-plate basalts. (b) Ha, Hauff et al. (2000a). (c) B,
724
Buchs et al. (2010).
725
32
726
Figure 8 (a) General petrogenetic paths followed by MORB and VAB and fields of MORB,
727
VAB and boninite (BON) in Y vs. Cr space and Y vs. Cr of (b) subduction initation rule
728
ophiolite basalts, (c) basalts of the Sona-Azuero, Golfito and Chagres-Bayano complexes
729
interpreted as oceanic plateau and our Groups I and II Azuero basalts and (d) arc basalts of the
730
Sona-Azuero, Chagres-Bayano and Golfito complexes and our Group III Azuero basalts (only,
731
no intrusives). All Central American samples are basalts only. The solid and dashed curves
732
represent incremental batch melting trends and source 1 (S1) and source 2 (S2) represent source
733
compositions from Murton (1989). S1 represents a calculated plagioclase lherzolite MORB
734
source comprised of 0.60 oli + 0.20 opx + 0.10 cpx and S2 represents the residue (residual
735
source) subsequent to 20% melt extraction from S1. The fields of MORB, VAB and within-plate
736
basalt (WPB) are from Pearce (1982) and the boninite (BON) field is from Dilek et al. (2007).
737
Abbreviations in: (b) bas-and, basaltic andesite; LU, lower unit; SIRO, subduction initiation rule
738
ophiolite; UU, upper unit. (c) enr, enriched; Ha, Hauff et al. (2000).
739 740
Figure 9 Ti/V (Shervais 1982) of (a) subduction initiation rule ophiolite (SIRO) basalts, (b)
741
Sona-Azuero, Golfito and Chagres-Bayano basalts interpreted as oceanic plateau and our Groups
742
I and II Azuero basalts and (c) Sona-Azuero, Golfito and Chagres-Bayano basalts interpreted as
743
arc and our Group III basalts. All Central American samples are basalts only (i.e., no intrusives).
744
Note that two of our enriched Sona-Azuero basalts fall off the plot to the right with 4.7-4.8 wt. %
745
TiO2 (~28000-29000 ppm Ti) and ~460 ppm V (Ti/V of ~53) in Fig. 9b, e within the within-
746
plate basalt field. The Izu Bonin Mariana FAB and Expedition 352 FAB are from Reagan et al.
747
(2010) and Shervais et al. (2019), respectively and the Expedition 351 FA-like basalts are from
33
748
Arculus et al. (2015). Abbreviations in: (a) BON, boninite; LU, lower unit, UU, upper unit. (b)
749
alk, alkaline; Ha, Hauff et al. (2000a); LU, UU, as in (a). (c) B, Buchs et al. (2010).
750 751
Figure 10a Chemotemporal relations between the Golfito, Sona-Azuero and Chagres-Bayano as
752
shown on the ‘synthetic tectonostratigraphic chart’ of Buchs et al. (2010). Unit acronyms (from
753
left to right): AP, Azuero (oceanic) Plateau (89-85 Ma); GA, Golfito Arc (75-66 Ma); SA, Sona-
754
Azuero Arc (75-39 Ma); CBA, Chagres-Bayano Arc (70-39 Ma). (b) Stratigraphy of an Azueru
755
section (the Rio Quema Formation) reproduced from Corral et al. (2011) which provides
756
relations between the Azuero oceanic ‘plateau’, proto-arc and arc. Other acronyms: PISI, plume-
757
induced subduction initiation. References for ages are provided in Table 1.
758 759
Figure 11 Two alternative tectonic models to reconcile subduction initiation at the Central
760
American Volcanic Arc system. (a) The earliest stages of subduction initiation were catalyzed by
761
plume emplacement (modified from Whattam and Stern 2015a). This event began to first form
762
the Sona-Azuero complex which was then followed by incipient formation of the Chagres-
763
Bayano complex, subsequent to slab rollback and hinge retreat. (b) Collision-induced subduction
764
initiation (see Buchs et al. 2010 for details) and subsequent arc construction upon a ~10 m.y.
765
oceanic plateau (based on the ideas of Buchs et al. 2010 and Wegner et al. 2011).
766 767
Table captions
768
Table 1 Age constraints for the Golfito, Sona-Azuero and Chagres-Bayano complexes.
769
Abbreviations: A, arc; CR, Costa Rica; PA, proto-arc; PAN, Panama; OP, oceanic plateau.
770
Superscripts: a, Azuero Arc includes the proto-arc of Buchs et al. (2010); b, is to indicate that
34
771
this refers to a restricted segment of the Azuero Arc (the Río Quema Formation, Corral et al.,
772
2011); c, Golfito also consists of basaltic trachyandesite and andesite according to Buchs et al.
773
(2010) and references therein; a TAS plot (not shown) does suggest this, however a plot of Nb/Y
774
vs. Zr/Ti (not shown) suggests that designation of alkaline affinities is the result of alteration
775
(addition of alkalis); d, similarly, the Nb/Y vs. Zr/Ti plot suggests no alkaline affinities; e,
776
Wegner et al. (2011) suggested alteration could account for the high K2O contents of some rock
777
types which is confirmed by a Nb/Y vs. Zr/Ti plot; f and g, interpreted by Hauff et al. (2000) and
778
Buchs et al. (2010) as oceanic plateau and proto-arc/arc, respectively.
779 780
Table 2 Whole rock ICP-AES analyzes of Groups I, II, and III Azuero lavas from this study.
781
Abbreviation: bas-and, basaltic-andesite.
782 783
Table 3 Mean and range of La/Yb normalized to chondrite ([La/Yb]CN) of our Group I, II and III
784
Sona-Azuero lavas vs. Sona-Azuero lavas of other studies (Buchs et al. 2010; Wegner et al.
785
2011). Chondrite abundances normalized to are from Sun and McDonough (1989).
786 787
References
788
Alabaster T., Pearce, J.A. and Malpas, J., 1982. The volcanic stratigraphy and petrogenesis of the
789
Oman ophiolite complex. Contributions to Mineralogy and Petrology 81, 168-183.
790
Allègre, C.J., Hart, S.R. and Minster, J.F., 1983. Chemical structure and evolution of the mantle
791
and continents determined by inversion of Nd and Sr idotopic data, II. Numerical experiments
792
and discussion. Earth and Planetary Science Letters 66, 191-213.
35
793
Allibon, J., Monjoie, P., Lapierre, H., Jaillard, E., Bussy, F., Bosch, D. and Senebier, F., 2008.
794
The contribution of the young Cretaceous Caribbean Oceanic Plateau to the genesis of late
795
Cretaceous arc magmatism in the Cordillera Occidental of Ecuador. Journal of South
796
American Earth Sciences 26, 355-368.
797 798 799 800
Arculus, R.J., 2003, Use and abuse of the terms calcalkaline and calcalkalic. Journal of Petrology 44, 929-935. Arculus, R.J. and the 29 others of IODP Expedition 351, 2015. A record of subduction initiation in the Izu-Bonin-Mariana arc. Nature Geoscience, http://dx.doi.org/10.1038/ngeo2515.
801
Barat, F., Mercier de Lépinay, B., Sosson, M., Müller, C., Baumgartner, P., and Baumgartner-
802
Mora, C., 2014. Transition from the Farallon Plate subduction to collision between South and
803
Central America: Geological evolution of the Panama isthmus. Tectonophysics 622, 145-167.
804
Beccaluva, L., Coltori, M., Giunta, G., and Siena, F., 2004. Tethyan vs. Cordilleran ophiolites: A
805
reappraisal of distinctive tectono-magmatic features of supra-subduction complexes in relation
806
to the subduction mode. Tectonophysics 39, 163-174.
807 808
Brounce, M., Kelley, K.A., Cotrell, E., and Reagan, M.K., 2015. Temporal evolution of mantle wedge oxygen fugacity during subduction initiation. Geology 43, 775-778.
809
Buchs, D.M., Arculus, R.J., Baumgartner, P.O., Baumgartner-Mora, C., and Ulianov, A., 2010.
810
Late Cretaceous arc development on the SW margin of the Caribbean Plate: Insights from
811
the Golfito, Costa Rica, and Azuero, Panama, complex. Geochemistry, Geophysics,
812
Geosystems, 11. doi:10.1029/2009GC002901.
813
Buchs, D.M., Baumgartner, P.O., Baumgartner‐Mora, C., Bandini, A.N., Jackett, S.‐J., Diserens,
814
M.‐O., and Stucki, J., 2009. Late Cretaceous to Miocene seamount accretion and mélange
815
formation in the Osa and Burica peninsulas (southern Costa Rica): Episodic growth of a 36
816
convergent margin, in The Origin and Evolution of the Caribbean Plate, edited by K. H.
817
James, M. A. Lorente, and J. L. Pindell, Geological Society Special Publication, v. 328, pp.
818
411–456.
819 820 821 822 823 824
Burov, E., and Cloetingh, S., 2010, Plume-like upper mantle instabilities drive subduction initiation: Geophysical Research Letters, v. 37, L03309, doi:10.1029/2009GL041535. Chauvel, C., and Blichert-Toft, J., 2001. A Hafnium isotope and trace element perspective on melting of the depleted mantle. Earth and Planetary Science Letters 190, 131-151. Cooper, P.A. and Taylor, B., 1985. Polarity reversal in the Solomon Islands arc, Nature 314, 428–430.
825
Corral, I., Griera, A., Gómez-Gras, D., Corbella, M., Canals, À., Pineda-Falconett, M., and
826
Cardellach, E., 2011. Geology of the Cerra Quema Au-Cu deposit (Azuero Peninsula,
827
Panama): Geologica Acta, 9(3-4), 481-498, doi : 1 0 . 1 3 4 4 / 1 0 5 . 0 0 0 0 0 1 7 4 2.
828
DeBari, S.M., Taylor, B., Spencer, K. and Fujioka, K., 1999. A trapped Philippine Sea plate
829
origin for MORB from the inner slope of the Izu-Bonin trench. Earth and Planetary Science
830
Letters 174, 183–197.
831 832
Dilek, Y., Furnes, H. and Shallo, M., 2007. Suprasubduction zone ophiolite formation along the periphery of Mesozoic Gondwana. Gondwana Research 11, 453-475.
833
Dilek Y., Furnes, H. and Shallo M., 2008. Geochemistry of the Jurassic Mirdita Ophiolite
834
(Albania) and the MORB to SSZ evolution of a marginal basin oceanic crust. Lithos 100,
835
174-209.
37
836
Elliot, T., Plank, T., Zindler, A., White, W., and Bourdon, B., 1997. Element transport from slab
837
to volcanic front at the Mariana arc. Journal of Geophysical Research 102, 14991–15019.
838
doi:10.1029/97JB00788.
839
Gazel, E., Carr, M.J., Hoernle, K., Feigenson, M.D., Hauff, F., Szymanski, D., and Van Den
840
Bogaard, P., 2009. The Galapagos-OIB signature in southern Central America: Mantle re-
841
fertilization
842
Q02S11,doi:10.1029/2008GC002246.
by
arc-hotspot
interaction.
Geochemistry,
Geophysics,
Geosystems,
843
Gazel, E., Hayes, J.L., Hoernle, K., Kelemen, P., Everson, E., Holbrook, W.S., Hauff, F., Van
844
Den Bogaard, P., Vance, E.A., Chu, S., Calvert, A.J., Carr, M.J., and Yogodzinski, G.M.,
845
2015. Continental Crust generated in oceanic arcs. Nature Geosciences 8, 321-327,
846
doi:10.1038/ngeo2392.
847
Geldmacher, J., Hanan, B. B., Blichert-Toft, J., Haarp, K., Hoernle, K., and Hauff, F., 2001.
848
Hafnium isotope variations in volcanic rocks from the Caribbean Large Igneous Province
849
and
850
doi:10.1029/2002GC000477.
Galápagos
hotspot
tracks.
Geochemistry,
Geophysics,
Geosystems
4,
851
Geldmacher, J., Hoernle, K., Van Den Bogaard, P., Hauff, F., and Klügel, A., 2008. Age and
852
geochemistry of the Central American forearc basement (DSDP Leg 67 and 84): Insights
853
into Mesozoic arc volcanism and seamount accretion on the fringe of the Caribbean LIP.
854
Journal of Petrology 49, 1781-1815, doi:10.1093/petrology/egn046.
855 856
Geldmacher, J., Heydolph, K., Murphy, D.T., Romanova, I., Mahoney, J. J. and K. Hoernle, 2012. American Geophysical Union, Fall Meeting 2012, abstract #DI51A-2339.
38
857
Gerya, T., Stern, R.J., Baes, M.. Sobolev, S.V. and Whattam. S.A., 2015. Plate tectonics on the
858
Earth
859
doi:10.1038/nature15752.
860 861 862 863
triggered
by
plume-induced
subduction
initiation:
Nature
527,
221-225,
Giudice, D. and Recchi, G., 1969. Geología del area del projecto minero de Azuero. Ejectora, informe tecnico. 48 pp., Gobierno de la Repub. de Panamá, Panama City. Goldstein, S. L., O’Nions, R.K., and Hamilton, P.J., 1988. Decoupled evolution of Nd and Sr isotopes in the continental crust. Nature 336, 733-738.
864
Hall, C.E., Gurnis, Sdrolias, M., Lavier, L.L. and Muellar R.D., 2003. Catastrophic initiation of
865
subduction following forced convergence across fracture zones, Earth and Planetary Science
866
Letters 212 15–30.
867
Hauff, F., Hoernle, K., Schmincke, H. –U., and Werner, R., 1997. A Mid-Cretaceous origin for
868
the Galapagos hotspot: Volcanological, petrological and geochemical evidence from Costa
869
Rican oceanic crustal fragments. Geol. Rundsch. 86, 141-155.
870
Hauff, F., Hoernle, K., and Schmidt A., 2003. Sr–Nd–Pb composition of Mesozoic Pacific
871
oceanic crust (site 1149 and 801, ODP Leg 185): Implications for alteration of ocean crust
872
and the input into the Izu–Bonin–Mariana subduction system. Geochemistry, Geophysics,
873
and Geosystems 4 (8).
874
Hauff, F., Hoernle, K.A., van den Bogaard, P., Alvarado, G.E. and Garbe‐Schonberg, D., 2000.
875
Age and geochemistry of basaltic complexes in western Costa Rica: Contributions to the
876
geotectonic evolution of Central America. Geochemistry, Geophysics, Geosystems 1, Paper
877
No. 1999GC000020.
39
878
Hawkesworth, C.J., Turner, S.P., McDermott, F., Peate, D.W. and van Calsteren, P., 1997. U-Th
879
isotopes in arc magmas: Implications for element transfer from the subducted crust. Science,
880
v. 276, p. 551-555.
881 882
Hoernle, K., Hauff, F., and van den Bogaard P., 2004. 70 m.y. history (139-69 Ma) for the Caribbean large igneous province. Geology 32, 697-700.
883
Hoernle, K., van den Bogaard, P., Werner, R., Lissinna, B., Hauff, F., Alvarado, G., and Garbe‐
884
Schonberg, D., 2002. Missing history (16–71 Ma) of the Galapagos hotspot: Implications for
885
the tectonic and biological evolution of the Americas. Geology 30, 795–798,
886
doi:10.1130/0091-7613(2002).
887
Ingle, S., Mahoney, J.J., Sato, H., Coffin, M.F., Kimura, J. –I., N. Hirano, N., and Nakanishi, M.,
888
2007. Depleted mantle wedge and sediment fingerprint in unusual basalts from the Manihiki
889
Plateau, central Pacific Ocean. Geology 35(7), 595-598, doi: 10.1130/G23741A.1.
890 891
Irvine, T.N., and Baragar, W.R.A., 1971. A guide to the chemical classification of the common volcanic rocks: Canadian Journal of Earth Sciences 8, 523-548, doi:10.1139/e71-055.
892
Ishizuka, O., Kimura, J.-I., Li, Yi B., Stern, R.J., Reagan, M.K., Taylor, R.N., Ohara, Y.,
893
Bloomer, S.H., Ishii, T., Hargrove III, U.S. and Haraguchi, S., 2006. Early stages in the
894
evolution of the Izu-Bonin arc volcanism: New age, chemical and isotopic constraints. Earth
895
and Planetary Science Letters 250, 385-401.
896
Ishizuka, O., Tani, K., Reagan, M.K., Kanayama, K., Umino, S., Harigane, Y., Sakamoto, I.,
897
Miyajima, Y., Yuasa, M., and Dunkley, D.J., 2011. The timescales of subduction initiation
898
and subsequent evolution of an oceanic island arc. Earth and Planetary Science Letters 306,
899
229-240.
40
900
Kerr, A.C., Marriner, G.F., Tarney, J., Nivia, A., Saunders, A.D., Thirlwall, M. J., and Sinton,
901
C.W., 1997. Cretaceous basaltic terranes in western Colombia: Elemental, chronological and
902
Sr–Nd isotopic constraints on petrogenesis. Journal of Petrology 38, 677-702.
903
Kerr, A.C., Tarney, J. Kempton, P.D., Spadea, P., Nivia, A., Marriner, G.F., and Duncan, R.A.,
904
2002. Pervasive mantle plume head heterogeneity: Evidence from the late Cretaceous
905
Caribbean-Colombian oceanic plateau. Journal of Geophysical Research 107(B7), doi:
906
10.1029/2001JB000790.
907 908
Kerr, A.C., White, R.V., and Saunders, A.D., 2000. LIP reading: Recognizing oceanic plateaux in the geological record. Journal of Petrology 41(7), 1041-1056.
909
Kerr, A.C., White, R.V., Thompson, P.M., Tamez, J., and Saunders, A.O., 2003. In: Bartolini,
910
C., Bufler, R.T., Blickwede, J. (Eds.), The Circum-Gulf of Mexico and the CaribbeanL
911
Hydrocarbon, habitats, basin formation and plate tectonics: American Associate of
912
petroleum Geologists (MPG) Memoir 79, pp. 126-188.
913
Kesler, S. E., Sutter, J.F., Issigonis, M.J., Jones, L.M., and Walker, R.L., 1977. Evolution of
914
porphyry copper mineralization in an oceanic island Panama. Economic Geology 72, 1142-
915
1153, doi:10.2113/gsecongeo.72. 6.1142.
916
Kolarsky, R.A., Mann, P., Monechi, S., Meyerhoff, H.D., and Pessagno Jr., E.A., 1995.
917
Stratigraphic development of southwestern Panama as determined from integration of
918
marine seismic data and onshore geology, in Geologic and Tectonic Development of the
919
Caribbean Plate Boundary in Southern Central America, edited by P. Mann, Special Paper,
920
Geological Society of America, v. 295, pp. 159-200.
921 922
Lissinna, B., Hoernle, K., and van den Bogaard, P., 2002. Northern migration of arc volcanism in western Panama: Evidence for subduction erosion? Eos Transactions, AGU 83, 1463–1464. 41
923
Lissinna, B., 2005. A profile though the Central American Landbridge in western Panama: 115
924
Ma interplay between the Galápagos Hotspot and the Central American Subduction Zone.
925
Ph.D. thesis, Christian-Albrechts University, Kiel, Germany, 102 pp.
926
Lissinna, B., Hoernle, K., Hauff, F., van den Bogaard, P., and Sadofsky, S., 2006. The
927
Panamanian island arc and Galápagos hotspot: A case study for the long-term evolution of
928
arc/hotspot interaction. Geophysical Research Abstracts 8, 05106.
929
Mamberti, M., Lapierre, H., Bosch, D., Jaillard, E., Ethien, R., Hernandez, J., and Polvé, M.,
930
2003. Accreted fragments of the Caribbean-Colombian Plateau in Ecuador. Lithos 66, 173-
931
199.
932 933 934 935
Marques, F.O., Cabral, F.R., Gerya, T.V., Zhu, G. and May, D.A., 2014. Subduction initiates at straight passive margins. Geology 42, 331-334. https://doi.org/10.1130/G35246.1. Matusomoto, T. and Tomoda, Y., 1983. Numerical simulation of the initiation of subduction at the fracture. Journal of Physics of the Earth 31, 183-194.
936
Maury, R.C., Defant, M.J., Bellon, H,. de Boer, Z.J., Stewart, R.H., and Cotten, J., 1995. Early
937
Tertiary arc volcanics from eastern Panama, in Geologic and Tectonic Development of the
938
Caribbean Plate Boundary in Southern Central America, edited by P. Mann, Specical Paper
939
of the Geological Society of America, 295, 29–34.
940
McCulloch, M.T. and Black, L.P., 1984. Sm–Nd isotopic systematics of Endreby land granulites
941
and
942
Planetary Science Letters 71, 46-58.
943 944
evidence for
the redistribution of Sm and Nd during metamorphism. Earth and
McCulloch, M.T. and Chappell, B.W., 1982. Nd isotopic charcteristics of S- and I-type granites. Earth and Planetary Science Letters 58, 51-64.
42
945 946
McDonough, W.F. and Sun, S.-S., 1995. The composition of the Earth, Chemical Geology 120, 223-253.
947
Meffre, S., Falloon, T. J. Crawford, T. J., Hoernle, K., Hauff, F. Duncan, R. A., Bloomer, S. H.
948
and Wright, D. J., 2012. Basalts erupted along the Tongan fore arc during subduction
949
initiation: Evidence from geochronology of dredged rocks from the Tonga fore arc and
950
trench.
951
doi:10.1029/2012GC004335.Metti, A., and Recchi, G., 1976. Geología de la Peninsula de
952
Soná e Isla de Coiba, in Memoria Segundo Congreso Latinoamericano de Geología, v. 2, pp.
953
541–553, Sucre, Caracas.
954 955
Geochemistry,
Geophysics,
Geosystems,
13,
Q12003,
Miyashiro, A., 1974. Volcanic rock series in island arcs and active continental margins. American Journal of Science 274, 321-355.
956
Montes, C., Bayona, G.A., Cardona, A., Buchs, D.M., Silva, C.A., Morón, S., Hoyos, N.,
957
Ramírez, D.A., Jaramillo, C.A., and Valencia, V., 2012a. Arc-continent collision and
958
orocline formation: Closing of the Central American Seaway. Journal of Geophysical
959
Research 117, B04105. doi.org/10.1029/2011JB008959.
960
Montes, C., Cardona, A., McFadden, R., Morón, S.E., Silva, C.A., Restrepo-Morono, S.,
961
Ramírez, D.A. Hoyos, N., Wilson, J., Farris, D., Bayona, G.A., Jaramillo, C.A., Valencia,
962
V., Bryan, J., and Flores, J.A., 2012b. Evidence for middle Eocene and younger land
963
emergence in central Panama: Implications for isthmus closure. Geological Society of
964
America Bulletin, doi:10.1130/B30528.1.
965 966 967
Nakamura, N., 1974. Determination of REE, Ba, Fe, Mg, Na and K in carbonaceous and ordinary meteorites. Geochimica et Cosmochimica Acta 38, 757-775. Pandey, D.K., Pandey, A. and Whattam, S.A., 2019. Relict subduction initiation along a passive 43
968
margin
969
https://doi.org/10.1038/s41467-019-10227-8.
in
the
northwest
Indian
Ocean.
Nature
Communications
10:2248,
970
Pearce, J.A., 1983. Role of the subcontinental lithosphere in magma genesis at active continental
971
margins, in Continental basalts and mantle xenoliths, edited by C. J. Hawkesworth and M.
972
J. Norry, pp. 230-249, Nantwich, Shiva.
973
Pearce, J.A., 2003. Supra-subduction zone ophiolites: The search for modern analogues, in
974
Ophiolite concept and the evolution of geological thought, edited by Y. Dilek, and S.
975
Newcomb, S., Geological Society of America Special Paper 373, pp. 269-293.
976 977
Pearce, J.A., 2008. Geochemical fingerprinting of oceanic basalts with applications to ophiolite classification and the search for Archean oceanic crust. Lithos 100, p. 14-48.
978
Pearce J.A, Lippard S. J., and Roberts, S., 1984. Characteristics and tectonic significance of
979
supra-subduction zone ophiolites, in Geology of Marginal Basins, edited by P. Kokelaar and
980
Howels, Geological Society of London Special Publication 16, pp. 77-94.
981
Pearce, J.A., Stern, R.J., Bloomer, S.H., and Fryer, P., 2005. Geochemical mapping of the
982
Mariana arc-basin system: Implications for the nature and distribution of subduction
983
components. Geochemistry, Geophysics, Geosystems 67, doi:10.1029/2004GC000895.
984 985
Pearce, J.A. and Norry, N.J., 1979. Petrogenetic implications of Ti, Zr, Y, and Nb variations in volcanic rocks. Contributions to Mineralogy and Petrology 69, 33-49.
986
Peucat, J.J., Vidal, P., Bernard-Griffiths, J., and Condie, K.C., 1988. Sr, Nd and Pb isotopic
987
systematics in the Archean low- to high-grade transition zone of southern India: Syn
988
accretion vs. post-accretion granulites. Journal of Geology 97, 537-550.
44
989
Phinney, E.J., Mann, P., Coffin, M.F., and Shipley, T.H., 1999. Sequence stratigraphy, structure,
990
and tectonic history of the southwestern Ontong Java Plateau adjacent to the North Solomon
991
Trench and Solomon Islands arc, Journal of Geophysical Research 104 (B9) 20446–20449.
992
Pindell, J.L., and Kennan, L., 2009. Tectonic evolution of the Gulf of Mexico, Caribbean and
993
northern South America in the mantle reference frame: an update, in The origin and
994
evolution of the Caribbean Plate, edited by K. H. James, M. A. Lorente, and J. L. Pindell,
995
328, London, Geological Society, pp. 1-55.
996
Portnyagin, M., Bindeman, I., Hoernle, K., and Hauff, F., 2007. Geochemistry of primitive lavas
997
of the Central Kamchatka Depression: Magma generation at the edge of the Pacific Plate, in
998
Volcanism and Subduction: The Kamchatka region, edited by J. Eichelberger, E. Gordeev,
999
M. Kasahara, P. Izbekov, and J. Lees, J., pp. 199-239. doi: 10.1029/172GM16.
1000
Reagan, M.K., Ishizuka, O., Stern, R.J., Kelley, K.A., Ohara, Y., Blichert-Toft, J,. Bloomer, S.–
1001
H., Cash, J., Fryer, P., Hanan, B.B., Hickey-Vargas, R., Ishii, T., Kimura, J.I., Peate, D.W.,
1002
Rowe, M.C. and Woods, M., 2010. Fore-arc basalts and subduction initiation in the Izu-
1003
Bonin-Mariana
1004
doi:10.1029/2009GC002871.
system:
Geochemistry,
Geophysics,
Geosystems
11,
Q03X12.
1005
Reagan, M.K., Pearce, J.A., Petronotis, K., Almeev, R.R., Avery, A.J., Carvallo, C., Chapman,
1006
T., Christeson, G.L., Ferré, E.C., Godard, M., Heaton, D.E., Kirchenbaur, M., Kurz, W.,
1007
Kutterolf, S., Li, H., Li, Y., Mishibayashi, K., Morgan, S., Nelson, W.R., Prytulak, J.,
1008
Python, M., Robertson, A.H.F., Ryan, J.G., Sager, W.W., Sakuyama, T., Shervais, J.W.,
1009
Shimizu, K., and Whattam, S.A. (2017) Subduction initiation and ophiolite crust: New
1010
insights
1011
http://dx.doi.org/10.1080/00206814.2016.1276482.
from
IODP
drilling.
45
International
Geology
Review,
1012
Robertson, A., 2004. Development of concepts concerning the genesis and emplacement of
1013
Tethyan ophiolites in the Eastern Mediterranean and Oman regions. Earth Science Reviews
1014
66, 331-387.
1015
Saccani, E., and Photiades, A., 2004. Mid-ocean ridge and supra-subduction affinities in the
1016
Pindos Massif ophiolites (Greece): Implications for magma genesis in a proto-forearc
1017
setting. Lithos 73, 229-53.
1018 1019
Salters, V.J.M., and Stracke, A., 2004. Composition of the depleted mantle. Geochemistry, Geophysics, Geosystems 5, (5), Q05004. doi:10.1029/2003GC000597.
1020
Shafaii Moghadam, H., Corfu, F., and Stern, R.J., 2013, U-Pb zircon ages of Late Cretaceous
1021
Nain-Dehshir ophiolites, central Iran: Journal of the Geological Society, v. 170, p. 175–184,
1022
https://doi .org /10 .1144 /jgs2012 -066.
1023 1024 1025 1026
Shaw, D.M., 1970. Trace element fractionation during anataxis. Geochimica et Cosmochimica Acta 34, 237-243. Shervais, J.W., 1982. Ti-V plots and the petrogenesis of modern and ophiolitic lavas. Earth and Planetary Science Letters 59, 110-118.
1027
Shervais, J.W., ,Reagan, M., , Haugen, E., , Almeev, R.R., , Pearce, J.A., , Prytulak, J., Ryan,
1028
J.G., , Whattam, S.A.,Godard, M., , Chapman, T., , Li, H., , Kurz, W.,, Nelson, W.R.,,
1029
Heaton, D.,, Kirchenbaur, M.Shimizu, K.,, Sakuyama, T.,Li, Y. , and Vetter, S.K., 2019.
1030
Magmatic Response to Subduction Initiation: Part 1. Fore-arc basalts of the Izu-Bonin Arc
1031
from IODP Expedition 352. Geochemistry, Geophysics, Geosystems, 20. https://doi.
1032
org/10.1029/2018GC007731
46
1033
Sinton, C.W., Duncan, R.A., and Denyer, P., 1997. Nicoya Peninsula, Costa Rica: A single suite
1034
of Caribbean oceanic plateau magmas. Journal of Geophysical Research 102, 15,507-15,520,
1035
doi:10.1029/97JB00681.
1036 1037
Stern, R.J., 2004. Subduction initiation: spontaneous and induced. Earth and Planetary Science Letters 226, 275-292.
1038
Stern, R.J. and Bloomer, S.H., 1992. Subduction zone infancy: examples from the Eocene Izu–
1039
Bonin–Mariana and Jurassic California. Geological Society of America Bulletin 104, 1621-
1040
1636.
1041
Stern, R.J. and Dumitru, T.A., 2019. Eocene initiation of the Cascadia subduction zone: A
1042
second
1043
https://doi.org/10.1130/GES02050.1.
1044 1045
example
of
plume-induced
subduction
initiation?
Geosphere.
Stern, R.J. and Gerya, T., 2018. Subduction initiation in nature and models: A review. Tectonophysics 746, 173-198.
1046
Stern, R.J., Reagan, M.,Ishizuka, O., Ohara, Y., and Whattam, S., 2012. To understand
1047
subduction initiation, study forearc crust: To understand forearc crust, study pphiolites.
1048
Lithosphere, doi: 10.1130/L183.1
1049
Sun, S.S. and McDonough, W.F., 1989. Chemical and isotopic systematic of oceanic basalts:
1050
Implications for mantle composition and processes, in Magmatism in the ocean basins,
1051
edited by A. D. Saunders and M. J. Norry, Geological Society of London Special Publication
1052
42, pp. 313-345.
1053
Tatsumi, Y. and Eggins, S., 1995. Subduction zone magmatism, Blackwell, London. 211 pp.
1054
Tomascak, P.B., Ryan, J.G., and Defant, M.J., 2000. Lithium isotope evidence for light element
1055
decoupling in the Panama subarc mantle. Geology 28, 507-510.
47
1056
Torró, L., Proenza, J.A., Marchesi, C., Garcia-Casco, A., Lewis, J.F., 2017. Petrogenesis of
1057
meta-volcanic rocks from the Maimón Formation (Dominican Republic): Geochemical
1058
record
1059
http://dx.doi.org/10.1016/j.lithos.2017.01.031.
of
the
nascent
Greater
Antilles
paleo-arc.
Lithos
278-281,
255-273,
1060
Tournon, J., Triboulet, C., and Azéma, J., 1989. Amphibolites from Panama: Anticlockwise P T
1061
paths from a pre-upper Cretaceous metamorphic basement in Isthmian Central America.
1062
Journal of Metamorphic Geology 7, 539–546, doi:10.1111/j.1525-1314.1989.tb00616.x.
1063 1064
Ueda, K., Gerya, T., and Sobolev, S.V., 2008, Subduction initiation by thermal-chemical plumes:
1065
Numerical studies: Physics of the Earth and Planetary Interiors, v. 171, p. 296–312,
1066
doi:10.1016/j .pepi.2008.06.032.
1067
Van der Lee, S., Regenauer-Lieb, K., and Yuen, D.A., 2008, The role of water in connecting past
1068
and future episodes of subduction: Earth and Planetary Science Letters, v. 273, p. 15–27,
1069
doi:10.1016/j .epsl.2008.04.041.
1070
Villagómez, D., Spikings, R., Magna, T., Kammer, A., Winkler, W., and Beltrán, A., 2011.
1071
Geochronology, geochemistry and tectonic evolution of the Western and Central cordilleras
1072
of Colombia. Lithos 125, 875-896.
1073
Wegner W., Wörner, G., Harmon R.S., and Jicha, B.R., 2011. Magmatic history and evolution of
1074
the Central American land bridge in Panama since Cretaceous times. Geological Society of
1075
America Bulletin 123, 703-734.
48
1076
Whattam, S.A., 2019. Primitive magmas in the early Central American Volcanic Arc system
1077
generated by plume-induced subduction initiation. Frontiers in Earth Science 6, 114, doi:
1078
10.3389/feart.2018.00114.
1079
Whattam, S.A., Malpas, J., Ali, J.R., Smith, I.E.M., 2008. New SW Pacific tectonic model:
1080
cyclical intraoceanic magmatic arc construction and near coeval emplacement along the
1081
Australia–Pacific margin in the Cenozoic. Geochemistry, Geophysics, Geosystems 9.
1082
doi:10.1029/2007GC001710.
1083
Whattam, S.A., Malpas, J., Smith, I.E.M., Ali, J.R., 2006. Link between SSZ ophiolite
1084
formation, emplacement and arc inception, Northland New Zealand: U–Pb SHRIMP
1085
constraints; Cenozoic SW Pacific tectonic implications. Earth and Planetary Science Letters
1086
250, 606–632.
1087
Whattam, S.A., Montes, C., McFadden, R.R., Cardona, A., Ramirez, D., and Valencia, V., 2012.
1088
Age and origin of earliest adakitic-like magmatism in Panama: Implications for the tectonic
1089
evolution of the Panamanian magmatic arc system. Lithos 142-143, 226-244.
1090
Whattam, S.A. and Stern, R.J., 2011. The ‘subduction initiation rule’: A key for linking
1091
ophiolites, intra-oceanic forearcs, and subduction initiation. Contributions to Mineralogy and
1092
Petrology 62, 1031-1045. doi: 10.1007/s00410-011-0638-z.
1093
Whattam, S.A. and Stern, R.J., 2015a. Late Cretaceous plume-induced subduction initiation
1094
along the southern margin of the Caribbean and NW South America: The first documented
1095
example with implications for the onset of plate tectonics. Gondwana Research 27, 38-63.
1096
Whattam, S.A. and Stern, R.J., 2015b. Arc magmatic evolution and the construction of
1097
continental crust at the Central American Volcanic Arc system. International Geology
1098
Review, http://dx.doi.org/10.1080/00206814.2015.1103668.
49
1099
Wilson, J.T., 1966. Did the Atlantic close and re-open? Nature 211, 676-681.
1100
Winchester, J.A. and Floyd, P.A., 1977. Geochemical discrimination of different magma series
1101 1102 1103
and their differentiation products using immobile elements: Chemical Geology 20, 325-343. Workman, R.K. and Hart, S.R., 2005. Major and trace element composition of the depleted MORB mantle (DMM): Earth and Planetary Science Letters 231(1-2), 53-72.
1104
Wörner, G., Harmon, R.S., and Wegner, W., 2009. Geochemical evolution of igneous rocks and
1105
changing magma sources during the evolution and closure of the Central American
1106
landbridge, in Backbone of the Americas: Shallow subduction, plateau uplift and ridge and
1107
terrane collision, edited by S. W. Kay, V. A. Ramos and W. Dickinson, Geological Society
1108
of America Memoir 204, pp.183-196.
1109
Zhou, X., Li, Z-H., Gerya, T.V., Stern, R.J., Xu, Z. and Zhang, J., 2018. Subduction initiation
1110
dynamics along a transform fault control trench curvature and ophiolite ages. Geology 46,
1111
607-610. https://doi.org/10.1130/G40154.1.
1112 1113
Zindler, A., and Hart, S.R., 1986. Chemical geodynamics. Annual Review of Earth and Planetary Sciences 14, 493-571.
50
Golfito Complex (i.e.,Proto-Arc, southern Costa Rica)
submarine
Azuero Complex (S. Panama) Azuero Plateau
submarine
Azuero Proto-Arc & Arc
Chagres-Bayano Arc (E. Panama)
Table 1 Whattam et al., 2019
submarine
?
subalkaline basalt, basaltic trachyandesite, trachyandesite
massive to thin pillowed lava flows
late Campanian to middle Maastrichian (~72-69 Ma)
biostratigraphic fossil ID
?
subalkaline basalt, (minor) gabbro
sheet flows, pillow basalts, rare gabbroic intrusives
Coniacian-Early Santonian (~89-85)
biostratigraphic fossil (radiolarite) ID
oceanic plateau
~1.7
tholeiitic basalt to rhyolite and their plutonic equilvalents (including enriched OIB-like varieties)
lavas flows, dykes, lava domes, large intrusive complexes
Late Cretaceous to Early Maastrichian 82-40
biostratigraphic fossil (foraminifera) ID;
proto-arc/arc
subalkaline basalt, (minor) gabbro basalt to rhyolite and their plutonic equilvalents
massive and pillowed lava flow-dyke complex with associated sub-volcanic plutonic rocks with overlying and intercalated volcaniclastic breccias
70-39
radiometric (40Ar/ 39Ar, LA-ICP-MS U-Pb on zircons)
~3
40
oceanic plateau and Dengo [1962]; Schmidt-Effing [1979]; Obando [1976]; DiMarco [1994]; DiMarco et al. [1995]; proto-arc/arc Hauff et al. [2000a]; Buchs et al. (2010)
Giudice and Recchi [1969]; Metti and Recchi [1976]; Kolarsky et al. [1995]; Hoernle et al. [2002]; Lissinna et al. [2002, 2006]; Lissinna [2005]; Buchs et al. [2009, 2010]; Wörner et al. [2009]; Corral et al. [2011]; Wegner et al. [2011]; Montes et al. [2012a]
39
radiometric ( Ar/ Ar, LA-ICP-MS U-Pb on zircons) arc
Maury et al. [1995]; Wörner et al. [2005, 2009]; Wegner et al. [2011]; Montes et al. [2012a, b]
Sample No. Lithology a Group
300070 basalt Group I
300071 basalt Group I
300075 basalt Group I
300076 basalt Group I
300077 basalt Group I
300078 basalt Group I
300079 basalt Group I
300081 basalt Group I
300082 basalt Group I
300082 (Rep) basalt Group I
300084 basalt Group I
300085 basalt Group I
LOI total
0.18 7.38 10.33 3.60 0.16 0.10 2.20 99.71
48.92 1.20 13.42 11.74 0.18 8.38 9.91 3.29 0.18 0.10 2.40 99.72
47.29 1.22 13.92 12.02 0.21 7.26 12.76 2.09 0.07 0.10 2.80 99.74
47.38 1.23 13.28 11.58 0.23 8.05 12.40 2.00 0.07 0.10 3.40 99.72
49.70 1.24 13.67 10.80 0.18 7.17 9.72 2.98 1.74 0.10 2.40 99.70
47.00 1.22 13.70 11.92 0.22 8.13 12.20 2.23 0.10 0.10 2.90 99.72
46.94 1.26 14.06 12.12 0.23 7.01 12.51 2.53 0.09 0.10 2.90 99.75
46.87 1.09 14.03 11.16 0.26 8.32 11.96 2.30 0.07 0.09 3.60 99.75
46.61 1.22 14.02 11.46 0.31 7.84 11.71 2.16 0.07 0.09 4.20 99.69
46.4 1.23 13.98 11.53 0.30 7.99 11.71 2.17 0.08 0.09 4.20 99.77
45.11 0.88 13.39 10.29 0.18 8.59 9.83 2.31 0.57 0.07 8.50 99.72
48.00 0.95 14.10 10.50 0.18 8.40 9.82 3.56 0.28 0.07 3.80 99.66
Mg#
56
59
55
58
57
58
54
60
58
58
62
61
48 316 46.4 138.1 67 2.4 204.8 21.6 58.7 3.9 <0.1 50 3.6
48 326 76.2 143.3 54 0.7 162.7 22.7 84.1 3.6 <0.1 14 3.2
50 331 69.6 142.8 61 0.6 118.1 22.6 61.8 3.7 <0.1 15 3.3
48 308 76.0 144.4 60 18.5 209.4 20.6 57.0 3.3 <0.1 204 2.8
51 336 69.4 140.1 49 1.1 155.4 22.4 58.2 3.2 <0.1 17 3.1
50 324 85.5 145.8 57 1.0 108.1 22.0 56.7 3.0 <0.1 17 3.0
48 289 65.8 128.3 50 0.7 142.2 18.7 66.1 2.9 <0.1 60 2.6
50 310 66.4 138.0 57 0.7 156.7 19.9 52.4 3.3 <0.1 27 2.8
50 314 NA NA NA 0.6 158.8 20.8 63.1 3.1 <0.1 31 2.9
46 275 51.5 109.7 49 1.1 134.9 16.0 41.8 3.2 <0.1 18 2.3
47 325 64.3 88.0 36 3.0 193.3 19.0 43.1 2.4 <0.1 151 2.5
major oxides (wt. %) SiO2 49.23 TiO2 1.22 Al2O3 13.55 Fe2O3 11.76 MnO MgO CaO Na2O K2O P2O5
trace elements (ppm) Sc 47 V 328 Ni 68.1 Cu 140.4 Zn 56 Rb 2.4 Sr 233.9 Y 21.9 Zr 59.2 Nb 3.3 Cs <0.1 Ba 37 3.2 La
Sample No. Lithology a Group
300086 basalt Group I
300088 basalt Group I
300089 basalt Group I
300091 basalt Group I
300072 basalt Group II
300073 basalt Group II
300074 basalt Group II
300083 basalt Group II
300066 andesite Group III
300067 basalt Group III
300068 basalt Group III
LOI total
0.20 8.10 11.78 2.39 0.14 0.10 2.40 99.69
47.93 1.25 16.75 11.04 0.20 6.40 9.48 3.29 0.67 0.11 2.60 99.72
48.78 1.76 13.72 13.18 0.21 6.23 10.80 2.74 0.17 0.18 2.00 99.77
47.04 1.26 13.86 11.74 0.21 7.55 12.64 1.80 0.04 0.10 3.50 99.74
44.77 4.55 11.78 16.96 0.23 5.52 8.34 3.46 0.46 0.44 3.20 99.71
43.97 4.69 11.82 16.68 0.21 5.46 11.06 2.15 0.51 0.45 2.70 99.70
46.95 2.76 13.34 14.18 0.22 6.51 9.82 3.25 0.33 0.24 2.10 99.70
45.00 3.64 13.09 13.37 0.17 5.62 9.46 3.86 0.92 0.43 4.20 99.76
56.28 1.14 15.37 9.05 0.17 3.46 6.37 3.94 0.99 0.31 2.80 99.88
48.11 0.57 20.30 8.79 0.15 4.91 10.80 3.23 0.47 0.07 2.40 99.80
47.06 0.51 18.78 10.11 0.19 7.24 12.99 1.63 0.10 0.05 1.10 99.76
Mg#
57
54
48
56
39
39
48
46
43
53
59
44 308 46.4 182.7 65 7.3 296.3 23.8 68.4 3.3 0.3 140 3.3
53 351 17.3 100.1 64 1.2 146.0 37.4 105.0 2.2 <0.1 71 4.2
48 309 73.6 135.4 52 0.4 172.0 21.4 63.3 3.8 <0.1 12 3.3
38 462 34.8 317.7 139 5.0 72.7 51.2 272.2 20.3 <0.1 51 17.4
38 463 34.6 323.7 107 7.1 155.9 53.4 281.5 20.9 <0.1 59 17.7
39 426 51.1 140.9 88 4.5 218.3 32.0 141.1 12.0 <0.1 92 9.8
28 336 56.0 25.1 75 16.5 473.0 29.5 237.6 32.7 <0.1 140 23.4
31 193 2.3 42.4 57 11.6 229.5 31.7 101.9 3.3 <0.1 397 10.9
33 241 16.3 151.5 40 7.2 373.7 11.6 25.5 1.2 0.3 159 2.6
43 296 9.9 221.6 17 2.1 360.7 10.7 11.0 0.8 0.1 72 2.3
major oxides (wt. %) SiO2 47.21 TiO2 1.22 Al2O3 14.03 Fe2O3 12.12 MnO MgO CaO Na2O K2O P2O5
trace elements (ppm) Sc 47 V 343 Ni 48.0 Cu 138.3 Zn 56 Rb 1.7 Sr 143.2 Y 23.0 Zr 59.1 Nb 3.6 Cs 0.1 Ba 60 3.2 La
Sample No. 300069 Lithology bas-and a Group Group III major oxides (wt. %) SiO2 54.30 TiO2 0.19 Al2O3 11.39 Fe2O3 11.49
300087 basalt Group III
300090 STD SO-18 STD SO-18 STD SO-18 STD SO-18 basalt Rep 1 Rep 2 Rep 3 Group III
LOI total
0.19 7.40 7.84 2.20 0.29 0.07 4.30 99.66
47.40 1.74 14.29 12.70 0.22 5.85 8.85 3.49 1.14 0.17 3.80 99.65
50.34 0.44 18.34 10.06 0.20 5.28 9.80 2.25 0.32 0.10 2.60 99.73
Mg#
56
48
51
47
47
47
47
44 336 19.2 107.8 75 14.5 336.9 32.9 100.2 2.5 <0.1 1100 5.0
41 267 9.1 118.1 35 8.2 764.3 13.5 25.6 0.5 0.1 219 3.4
26 192 46 NA NA 26.4 377.8 29 261.1 18.9 6.5 476 11.2
26 191 45 NA NA 26.4 380.7 29.7 264 19.1 6.7 484 11.4
25 205 46 NA NA 27.8 399 31.3 278.9 19.4 6.7 500 12.1
25 207 43 NA NA 28 413.7 31.4 282.7 20.5 7 513 12.2
MnO MgO CaO Na2O K2O P2O5
trace elements (ppm) Sc 59 V 338 Ni 20.3 Cu 71.7 Zn 128 Rb 3.2 Sr 164.8 Y 8.8 Zr 15.0 Nb 1.8 Cs <0.1 Ba 668 3.2 La
58.29 0.7 13.98 7.59 0.4 3.34 6.31 3.71 2.15 0.83 1.9 99.76
58.22 0.7 13.98 7.59 0.4 3.35 6.35 3.7 2.16 0.84 1.9 99.76
58.13 0.69 14.11 7.58 0.39 3.35 6.36 3.7 2.15 0.82 1.9 99.74
58.11 0.69 14.13 7.58 0.39 3.33 6.38 3.71 2.16 0.81 1.9 99.74
Ce
8.4
8.7
8.7
8.7
7.7
7.7
8.0
6.8
7.6
7.4
5.6
7.2
Ce
9.1
8.7
12.0
8.2
45.7
46.8
24.5
55
24.0
5.6
4.6
Ce
4.9
13.8
6.9
24.7
25.2
26.9
27.5
Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U
1.37 7.5 2.4 0.90 3.18 0.63 3.78 0.80 2.34 0.36 2.28 0.35 1.9 0.2 0.6 0.3 0.1
1.4 7.3 2.35 0.91 3.25 0.63 3.80 0.82 2.37 0.36 2.36 0.36 2 0.2 0.2 <0.2 0.1
1.42 7.1 2.51 0.97 3.39 0.64 3.76 0.82 2.44 0.36 2.30 0.36 2.3 0.2 0.6 0.3 0.1
1.41 7.4 2.47 0.92 3.40 0.64 3.86 0.83 2.41 0.35 2.33 0.35 2.0 0.3 0.2 0.2 0.2
1.32 6.5 2.19 0.90 2.97 0.59 3.77 0.79 2.30 0.35 2.35 0.34 2.0 0.2 0.2 0.4 0.1
1.35 7.5 2.35 0.92 3.24 0.62 3.90 0.82 2.40 0.37 2.30 0.33 1.8 0.2 0.3 <0.2 0.1
1.34 7.0 2.23 0.89 3.17 0.60 3.92 0.77 2.35 0.36 2.33 0.34 1.8 0.2 0.3 0.2 0.1
1.14 5.9 1.99 0.79 2.75 0.54 3.24 0.69 2.04 0.31 1.97 0.30 2.0 0.2 0.2 <0.2 <0.1
1.22 6.4 2.17 0.89 3.03 0.57 3.49 0.74 2.06 0.33 2.15 0.33 1.7 0.3 <0.1 <0.2 0.1
1.24 6.6 2.11 0.89 2.96 0.58 3.54 0.73 2.15 0.31 2.25 0.33 1.5 0.1 NA 0.2 0.1
0.97 5.1 1.62 0.63 2.27 0.44 2.71 0.59 1.69 0.26 1.84 0.26 1.2 0.1 0.1 0.2 <0.1
1.04 6.4 1.93 0.72 2.44 0.49 3.22 0.67 2.00 0.29 1.79 0.27 1.3 0.2 <0.1 <0.2 <0.1
Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U
1.42 7.1 2.14 0.99 3.14 0.61 3.66 0.85 2.62 0.34 2.29 0.34 1.8 0.2 0.3 <0.2 <0.1
1.39 7.4 2.41 0.92 3.37 0.65 4.22 0.89 2.68 0.39 2.55 0.39 1.9 0.2 0.2 0.4 0.1
2.13 12.6 4.00 1.41 5.44 1.04 6.42 1.38 4.11 0.62 4.03 0.59 3.1 0.1 0.8 0.3 <0.1
1.36 7.1 2.28 0.88 3.15 0.62 3.75 0.78 2.29 0.34 2.22 0.32 1.8 0.3 0.3 0.2 <0.1
6.87 34 9.23 2.83 10.75 1.76 10.06 1.84 5.02 0.68 4.36 0.60 7.7 1.5 1.1 1.5 0.5
7.31 34 9.74 2.91 10.94 1.83 10.05 1.97 5.22 0.71 4.65 0.62 8.5 1.3 1.1 1.3 0.6
3.77 18.1 5.33 1.85 6.21 1.06 6.07 1.20 3.24 0.47 3.07 0.42 4.0 0.8 0.6 0.8 0.4
7.27 33 7.21 2.39 7.47 1.11 6.00 1.05 2.80 0.37 2.39 0.33 6.3 2.1 2.1 1.8 0.5
3.51 16.4 4.28 1.37 5.23 0.89 5.63 1.16 3.43 0.51 3.51 0.50 3.1 0.2 1.8 1.9 0.6
0.89 4.2 1.29 0.57 1.74 0.33 2.02 0.44 1.32 0.21 1.31 0.20 0.9 <0.1 1.1 0.5 0.2
0.73 3.4 1.28 0.58 1.69 0.32 1.91 0.42 1.19 0.20 1.29 0.20 0.4 <0.1 1.3 <0.2 <0.1
Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U
0.90 4.2 1.05 0.35 1.24 0.22 1.26 0.28 0.85 0.15 1.05 0.18 0.5 <0.1 2.3 0.3 0.2
2.32 12.2 3.65 1.42 4.87 0.90 5.70 1.18 3.48 0.52 3.40 0.50 2.5 0.1 0.4 0.5 0.1
1.16 5.9 1.62 0.57 2.03 0.34 2.13 0.48 1.44 0.23 1.54 0.24 0.9 <0.1 1.3 0.4 0.2
3.15 13.1 2.73 0.81 2.77 0.48 2.74 0.58 1.71 0.27 1.63 0.27 8.8 6.8 NA 9.9 15.3
3.16 12.7 2.71 0.84 2.8 0.47 2.73 0.58 1.7 0.26 1.69 0.26 9 6.9 NA 9.9 15.4
3.27 13.6 2.84 0.84 2.95 0.5 2.86 0.6 1.71 0.27 1.73 0.26 9.2 6.9 NA 10.4 15.8
3.31 13.4 2.84 0.84 2.94 0.49 2.88 0.61 1.79 0.27 1.74 0.27 9.6 7.2 NA 10.1 16.1
a
Group refers to classification on the basis of petrographic characteristics (see Supplementary Document) and trace element chemistry (see Section 5.1 and Supplementary Document). Major element detection limits are about 0.001-0.4%. Other notes: Samples Group numbers in italics are samples with total oxide concentrations <96 wt. % and which were thus excluded from our chemical plots (see Section 4.3.2.). NA = Not analyzed. Rep = Replicate (analysis). STD = Standard. Table 2 Whattam et al., 2019
Table 3 Comparison of chondrite normalized La/Yb of Azuero samples with Sona-Azuero samples in the literature
Our data
Others
Group
mean [La/Yb]CN
range
n
mean [La/Yb]CN
range
n
I II III
0.96 2.63 1.43
0.75-1.09 2.29-2.86 1.28-1.58
12 3 3
0.92 2.39 1.43
0.63-1.15 1.96-2.81 0.76-3.21
36 3 17
Notes: The Group I of others comprise the Azuero and Sona 'CLIP' basalts of Wegner et al. [2011] and the Group I Azuero Plateau basalts of Buchs et al. [2010]; the Group II of others comprise the Group II basalts of Buchs et al. (2010); the Group III of others comprise arc basalts of Azuero, Sona and Coiba Island of Lissinna [2005] and the basalts of the proto-arc and arc of Buchs et al. [2010]. Chondrite abundances are from Sun and McDonough [1989].
Table 3 Whattam et al., 2019
100o W
o
o
90 W
70o W
80 W
NORTH AMERICAN PLATE Yucatan
60o W
N
Hispaniola
Cuba
Virgin Islands
Jamaica
CR 400 km
PR
GAA system
A s y s t em
CARIBBEAN PLATE
LA
Nicaragua
20o N
ean LIP ribb Ca Curacao
PAN
Fig. 1b
10oN
Venezuela
COCOS PLATE Colombia
a
Galapagos Islands
SOUTH AMERICAN PLATE
NAZCA PLATE
Caribbean Large Igneous Province (CLIP) PCFZ
08o N
PC
Pa n Fig. 2
Sona
Pearl Islands
Sona-Azuero Arc
Sona
CI
CB Arc
Majé
o
de
an
Cord
N
Chagre s-B ay
PANAMA Golfito GF Arc
b
CARIBBEAN PLATE
COSTA RICA 10 N o
o
0N
Bahia Pinia
Azuero Azuero
o
82 W
o
80 W
250 km
Figure 1 Whattam et al., 2019
a
Northwest
Southeast Caribbean coast
Pacific coast
eastern Panama
southern Costa Rica
southern Panama
Golfito Complex
Azuero Chagres-Bayano Complex Complex
1
PALEOGENE
Ma 40 50
70 80 90
Achiote Fm.
PALEOCENE Maastrichtian
L. CRETACEOUS
60
EOCENE
1
GA
Campanian Santonian Coniacian Turonian Cenomanian
100
CBA
~3 km
SAA
2
?
?
?
AP
AP
AP
? no sediment horizon or unconformity?
timing of PISI
“old plateau”
b NC
Rio Quema Formation Stratigraphic Section
calcarenite basaltic andesite dyke
v v
siltstone
Tonosi Formation meters
hemipelagic limestone
Azuero Arc
v
1500 +
+
+
+
+
+ v
v +
+
+
+
+
+
+
+
v
+
+
+
+
+
+
+
+
+
+
+
+
+
conglomerate
v +
+
+
+
+
+
+
+
+
+
+
+
+
dacite lava dome
+
+ + + + +
v v
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
crystal rich sandstone andesite basalt
v
+ + + + + + + + + + + + + + + +
v
+
v
+
+
v
+
v
v
+
v
+
v
v
+
v
+
+
v
+
+
+
v
+
Upper Unit +
v v v
+
v
+
+
v
+
+
v v v v v v
+
+
v
+
v
+
v v v v v v
+
+ + + + + + + + + +
+ + + + + + + + + +
+ + + + + + + + + + + + + + + + + + + +
v +
v
1000
turbidite
+
v
+ + + + + + + + + + + + + + + +
v
v v v v
v v v v v v v v
v v v v
v v v v
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
vv v v v v v v v
v
+
radiolarite
v
+
v
v v v v v v v v
500
v
pillow basalt
+
Limestone Unit + +
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ v
v v
v
v v
v
v v
v
v v
v
v v
v
v
0
+
v v v v v v
+ + +
Lower Unit
+
Proto-Arc Unit Azuero Igneous Basement (Oceanic Plateau)
Figure 10 Whattam et al. 2019
2
Golfito Formation Ocu Formation overlap sediments (forearc deposits) arc related igneous rocks & sediments proto-arc igneous rocks hemipelagic limestone oceanic plateau
a South
North zone of plumeweakened lithosphere BUOYANT
100-90 Ma: construction of CLIP Pacific
DENSE
thin oceanic crust
old thick lithosphere upwelling plume-modified asthenosphere
DENSE
mantle plume
85-75 Ma ?
lower, older Sona-Azuero OP & proto-arc/arc suite
thick OP crust of Azuero Plateau
Farallon Plate
proto-forearc spreading
stages of subduction initation
? thick OP crust of CLIP
Farallon Plate plu
me
flo
?
mantle plume(s)
90 Ma: plume-induced subduction initation
early
BUOYANT
89-85 Ma: construction of Azuero Plateau Pacific
thin oceanic crust
old thick lithosphere
b North
South
w
70 Ma partial melting of plume-contaminated MORB-like source subsequent to plume-induced subduction initiation (PISI); little to moderate fluid (shallow) subduction additions
plume-modified asthenosphere
75-73 Ma: collision-induced subduction initiation partial melting of plume-contaminated source ~ 10 m.y. subsequent to Azuero Plateau construction and subquent proto-arc and arc formation
upper, younger arc suite lower, older proto-arc suite Azuero Plateau
upper, younger Chagres-Bayano arc suite forearc spreading late hinge retreat
partial melting of an increasingly subduction-modified and deplteted source, via early shallow fluid flux hydration of mantle wedge & late (deep) sediment melt flux assisted melting subsequent to extraction of lavas and intrusives of the Sona-Azuero Complex
Figure 11 Whattam et al. 2019
o
15’
81 00’
45’
30’
o
15’
80 00’
50 km 300090 (GIII) o
8 45’
300066 (GIII) 300067 (GIII) 300091 (GI)
300068 (GIII)
300089 (GI)
300069 (GIII) 300088 (GI) 300071 (GI) 300070 (GI) 300072 (GI) 300073 (GII) 300076 (GI) 300079 (GI) 300078(GI) 300075(GI) 300077(GI) 300074(GII) 300086(I) 300087(GIII) 300081(GI) 300085(GI) 300082(GI) 300083(GII) 300084(GI)
30’
Playa Venado
Azuero peninsula
15’
Overlap sediments & alluvial deposits
CR GF Nic
C
oc
R os
Panama
CB
SA
id
ge
Azuero Accretionary Complex (Paleocene to Eocene) Accreted seamounts & oceanic islands Azuero Arc Group (Maastrichtian to Middle Eocene) Volcanic sequences & sediments & intrusives
Azuero
Azuero Proto-Arc Group (Late Campanian) Azuero Plateau (Albian? to Coniacian-Early Santonian)
Figure 2 Whattam et al. 2019
Sona-Azuero (89-85 Ma) others
Sona-Azuero this study
Grp I Grp II (enriched)
Grp I Grp II (enr)
Sona-Azuero Sona-Azuero this study (73-39 Ma) others Grp III Grp III proto-arc & arc
MORB depleted arcs
Golfito (Ha) (75-66 Ma) Chagres-Bayano (70-39 Ma) Golfito proto-arc (75-66 Ma, B)
Chagres-Bayano arc (70-39 Ma)
20
MgO wt.%
15
10
5
0 40
50
60
70
80
SiO2 wt.%
Figure 3 Whattam et al. 2019
Sona-Azuero (89-85 Ma) others
Sona-Azuero this study
Grp I Grp II (enriched)
Grp I Grp II (enr)
Sona-Azuero Sona-Azuero this study (73-39 Ma) others Grp III Grp III proto-arc & arc
MORB depleted arcs
Golfito (Ha) (75-66 Ma) Chagres-Bayano (70-39 Ma) Golfito proto-arc (75-66 Ma, B)
Chagres-Bayano arc (70-39 Ma)
5
25
a
b 20
Al2O3 wt.%
TiO2 wt.%
4
3
2
15
10
5
1
3
20
c
d K2O wt.%
FeOt wt.%
15
10
2
1
5
50
10
e
f Yb ppm
Ce ppm
40
30
20
5
10
300
100
g
h Zr ppm
Y ppm
80
60
40
200
100 20
0
0 0
5
10
15
20
MgO wt.%
Figure 4 Whattam et al. 2019
0
5
10
MgO wt.%
15
20
200
a
100
10
2
Sona-Azuero Group I 89-85 Ma plateau/CLIP (others) early hybrid PAR (this study)
200
b
100
10 Sona-Azuero Group II
2
89-85 Ma enriched plateau (others) enriched
200
c
Rock/Chondrites
100
10
Sona-Azuero Group III 73-39 Ma proto-arc & arc (others) later arc-dominated PAR (this study)
200 Sona-Azuero
100
CLIP & our Gr I basalts
d
arc & our Gr III basalts
10
Golfito Complex (75-66 Ma) interpreted as proto-arc interpreted as CLIP OP
2 200
e
100
10
2
Chagres-Bayano Complex interpreted as CLIP OP
200 100
Chagres-Bayano Complex interpreted as arc
f
10
1 La Ce Pr Nd
Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
Fig.5 Whattam et al. 2018
400
a
100
Sona-Azuero Group I 89-85 Ma plateau/CLIP (others) early hybrid PAR (this study)
10
1
.1 400
b
100
10
1 Sona-Azuero Group II
.1
89-85 Ma enriched plateau (others) enriched
400
c
100
Sona-Azuero Group III 73-39 Ma proto-arc & arc (others) later arc-dominated PAR (this study)
Rock/N-MORB
10
1
.1 400
d
100
Golfito Complex (75-66 Ma)
10
interpreted as proto-arc interpreted as CLIP OP
1
.1 400
e
100 Chagres-Bayano Complex
interpreted as CLIP OP
10
1
.1 400
f
100 Chagres-Bayano Complex interpreted as arc
10
1
.1 Rb Th Nb La Pb Sr Nd Sm Ti Y Lu Cs Ba U K Ce Pr P Zr Eu Dy Yb
Fig. 6 Whattam et al. 2019
20
a 84% SIRO LU
100% SIRO UU
10
WPB
EM
DM
MORB
VAB 1
SIRO lower unit (LU) basalt SIRO upper unit (UU) bas-and
20 Sona-Azuero (89-85 Ma) others
10
Sona-Azuero this study Grp I Grp II (enr)
b
Golfito (Ha) (75-66 Ma) Chagres-Bayano (70-39 Ma)
Zr/Y
Grp I Grp II (enriched)
1
20 10
Sona-Azuero Sona-Azuero (73-39 Ma) others this study Grp III Grp III proto-arc & arc
Golfito proto-arc (75-66 Ma, B)
Chagres-Bayano arc (70-39 Ma)
c
all tholeiitic basalts (from b)
1 .5 10
100
500
Zr ppm
Figure 7 Whattam et al. 2019
7000
S2
S1
40%
1000
40%
to S3
20%
20%
10%
relative crystallization sequence of plag & cpx MORB: plag cpx VAB: cpx plag 20%
10%
40%
BON
10%
MORB fra
cti
fr a c
100
t io n a t io
Petrogeneses of MORB & VAB
on
ati
on
VAB
n
a
7000 S2
S1
40%
1000
SIRO LU basalt SIRO UU bas-and
40%
to S3
20%
10%
10%
20%
20%
40%
10%
Cr ppm
100
b
40%
7000
S2
S1
40%
1000
40%
to S3
20%
20%
10%
10%
20%
Golfito OP (Ha) (75-66 Ma) Chagres-Bayano OP
40% 10%
100 Sona-Azuero OP (89-85 Ma) others
Sona-Azuero this study
Grp I Grp II (enr)
7000
Grp I Grp II (enr)
S2
S1
40%
1000
40%
to S3
20%
20%
10%
30%
c
40%
Sona-Azuero proto-arc & arc (73-39 Ma) others Grp III
Sona-Azuero this study Grp III
20% 40% 10%
100 Golfito proto-arc (75-66 Ma) Chagres-Bayano arc (70-39 Ma)
40%
d
10 3
10
100
Y ppm
Figure 8 Whattam et al. 2019
Comparison with SIRO lavas
Comparison with FAB
BON
BON
ARC
600
V Ti/
=2
Izu-Bonin Mariana FAB Exp. 352 FAB Exp. 351 FA-like basalts
Ti/ V
SIRO LU basalt SIRO UU bas-andesite
Ti/ V
800
=1
=1
0
0
1000
ARC
0
MORB
400
= Ti/V
V Ti/
=2
0
MORB
50
= Ti/V
WPB
200
50
WPB
a
d
Golfito OP (Ha) (75-66 Ma)
Grp I Grp II (enr)
Chagres-Bayano OP
BON
SIRO UU
600
ARC
SIRO LU
V Ti/
=2
0
MORB
400
= Ti/V
50
WPB
200
e
b =1
0
1000 Sona-Azuero proto-arc & arc (73-39 Ma) others Grp III
800
Ti/ V
V ppm
Sona-Azuero this study
Grp I Grp II (enriched)
Ti/ V
Sona-Azuero OP (89-85 Ma) others
800
=1
0
1000
Sona-Azuero this study Grp III
600
ARC V Ti/
BON
Golfito proto-arc (75-66 Ma, B)
=
20
MORB
Chagres-Bayano arc (70-39 Ma)
400
= Ti/V
50
WPB
200
Group I (from b)
0
5
10
15
f
c
0 20
25 0
Ti ppm/1000
Figure 9 Whattam et al. 2019
5
10
15
20
25
Highlights
• • •
Older Sona-Azuero complex is MORB-like Younger Chagres-Bayano complex is arc-like Early Central American forearc adheres to SIR
S1342-937X(19)30283-7
DOI:
https://doi.org/10.1016/j.gr.2019.10.002
Reference:
GR 2231
To appear in:
Gondwana Research
Received Date: 12 October 2018 Revised Date:
11 September 2019
Accepted Date: 2 October 2019
Please cite this article as: Whattam, S.A., Montes, C., Stern, R.J., Early Central American forearc follows the subduction initiation rule, Gondwana Research, https://doi.org/10.1016/j.gr.2019.10.002. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.
Comparison with SIRO lavas
Comparison with FAB
BON
BON
ARC
600
V Ti/
=2
Izu-Bonin Mariana FAB Exp. 352 FAB Exp. 351 FA-like basalts
Ti/ V
SIRO LU basalt SIRO UU bas-andesite
Ti/ V
800
=1
=1
0
0
1000
ARC
0
MORB
400
= Ti/V
V Ti/
=2
0
MORB
50
= Ti/V
WPB
200
50
WPB
a
d
Golfito OP (Ha) (75-66 Ma)
Grp I Grp II (enr)
Chagres-Bayano OP
BON
SIRO UU
600
ARC
SIRO LU
V Ti/
=2
0
MORB
400
= Ti/V
50
WPB
200
e
b =1
0
1000 Sona-Azuero proto-arc & arc (73-39 Ma) others Grp III
800
Ti/ V
V ppm
Sona-Azuero this study
Grp I Grp II (enriched)
Ti/ V
Sona-Azuero OP (89-85 Ma) others
800
=1
0
1000
Sona-Azuero this study Grp III
600
ARC V Ti/
BON
Golfito proto-arc (75-66 Ma, B)
=
20
MORB
Chagres-Bayano arc (70-39 Ma)
400
= Ti/V
50
WPB
200
Group I (from b)
0
5
10
15
f
c
0 20
25 0
5
10
Ti ppm/1000
Graphical Abstract Whattam et al. 2019
15
20
25
1
Early Central American forearc follows the subduction initiation rule
2 3 4 5
Scott A. Whattama, Camilo Montesb, and Robert J. Sternc
6 7 8 9 10 11 12 13 14 15 16 17
a
Department of Geosciences, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia; Department of Physics and Geology, Universidad del Norte, Barranquilla, Colombia c Department of Geosciences, University of Texas at Dallas, Richardson, TX, USA b
Words: 7752 excluding abstract, acknowledgements and appendices
CONTACT Scott A. Whattam [email protected]
18
Keywords: Central American forearc; subduction initiation; subduction initiation rule; forearc;
19
volcanic arc; ophiolite
1
20
Abstract
21
The “subduction initiation rule“ (SIR) (Whattam and Stern, 2011) advocates that proto-arc and
22
forearc complexes preserved in ophiolites and forearcs follow a predictable chemotemporal
23
and/or chemostratigraphic vertical progression. This chemotemporal evolution is defined by a
24
progression from bottom to top, from less to more depleted and slab-metasomatized sources.
25
This progression has been recently documented for other igneous suites associated with
26
subduction initiation. The Sona-Azuero forearc complex of southern Panama represents the
27
earliest magmatic arc activity at the Central American Volcanic Arc system. Comparison of new
28
and existing geochemical data for the circa 82-40 Ma Sona-Azuero Proto-Arc/Arc, its underlying
29
89-85 Ma “oceanic plateau” of SW Panama and the 72-69 Ma Golfito Proto-Arc of southern
30
Costa Rica with the 70-39 Ma Chagres-Bayano Arc of eastern Panama exhibits a chemotemporal
31
progression as described above and which follows the SIR. Sona-Azuero lavas are
32
predominantly MORB-like, whereas those of the younger Chagres-Bayano complex are mostly
33
VAB-like; lavas of the Golfito Proto-Arc typically show characteristics intermediate to that of
34
the Sona-Azuero and Chagres-Bayano proto-arc/arc complexes. On the basis of isotope evidence
35
as shown in other studies, lava types of all three complexes are clearly derived from a source
36
contaminated by the Caribbean Large Igneous Province plume; we term these “plume-
37
contaminated” forearc basalts and volcanic arc basalts, respectively. Apart from a plume-induced
38
subduction initiation origin for the Panamanian forearc, these insights suggest otherwise similar
39
petrogenetic origins and tectonic setting to lavas comprising earliest-formed forearc crust, and
40
most ophiolites, which follow the SIR.
41 42
1. Introduction
2
43 44
Ophiolites are ancient, on-land vestiges of oceanic lithosphere ascribed to seafloor spreading to
45
form a nascent proto-forearc in the earliest stages of subduction initiation. As shown by Whattam
46
and Stern (2011), Tethyan-type supra-subuduction zone ophiolites (Pearce et al., 1984; Pearce,
47
2003) of the Eastern- Mediterranean-Persian Gulf region (Robertson, 2004) exhibit similar
48
chemostratigraphic/chemotemporal progressions (changes in lava chemistry with time) as that of
49
forearc lithosphere of classic records of magmatism during subduction initiation, such as the Izu-
50
Bonin-Mariana intra-oceanic arc system (Reagan et al., 2010; Ishizuka et al., 2011). Tethyan-
51
type ophiolites for example, in Oman (Semail), Greece (Pindos) and Albania (Mirdita) formed
52
by initial eruption of MOR-like tholeiitic basalts (i.e., forearc basalts, FAB, Reagan et al., 2010)
53
followed by calc-alkaline volcanic arc basalt (VAB) or basaltic-andesite lavas and intrusives
54
characterized by large ion lithophile-enrichment and high-field strength element-depletion,
55
which were sometimes followed by latest-stage boninites (e.g., Alabaster et al., 1982; Saccani
56
and Photiades, 2004; Dilek et al., 2008). This exact sequence is not universal but a common
57
theme is emerging from global examples. Variations in subduction initiation progression is
58
revealed by the Troodos ophiolite which lacks MORB-like tholeiites and instead shows a
59
progression from earlier-formed basalt-andesite-dacite-rhyolite series backarc basin-like basalts
60
to boninites (Pearce and Robinson, 2010). At the Izu-Bonin-Mariana forearc, generation of
61
earliest MORB-like FAB during initial subduction initiation was followed by eruption of
62
boninites (Reagan et al., 2010; 2017); from this time forward igneous activity retreated to the site
63
of the long-term magmatic front. A similar chemotemporal progression has recently been
64
documented for the nascent stages of Greater Antilles Arc system construction (Torró et al.,
65
2017). This chemotemporal progression reflects complementary magmatic and mantle responses
3
66
(MORB VAB affinities and fertile lherzolite depleted harzburgite, respectively) during
67
subduction initiation and encapsulates the ‘subduction initiation rule’ (SIR) of Whattam and
68
Stern (2011). The SIR stipulates that the composition of subduction initiation lavas change from
69
less to more high-field strength depleted and from less to more slab-metasomatized with time,
70
i.e., lavas change from MORB-like to VAB-like over the course of subduction initiation, and this
71
rule links forearcs formed during subduction initiation and ophiolites. This chemotemporal
72
progression evolution reflects formation of proto-forearc/forearc lithosphere as a result of mantle
73
melting that is increasingly influenced by enrichments from the sinking slab during subduction
74
initiation. Mid-oceanic ridge-like FAB lavas are the consequence of decompression melting of
75
upwelling asthenosphere and mark the initial magmatic manifestation of subduction initiation.
76
Contribution of fluids from dehydrating oceanic crust and sediments on the sinking slab is minor
77
during the early stages of subduction initiation, but continuous melting results in a depleted,
78
harzburgitic residue. This residue is increasingly metasomatized by fluids from the sinking slab
79
and later partial melting of this residue yields ‘typical’ suprasubduction-zone-like lavas with
80
high-field strength element depletions and large ion element enrichments in the latter stages of
81
subduction initiation. Recognizing the SIR has profound implications for the development of
82
robust tectonic plate reconstructions and interpreting and understanding forearc formation during
83
subduction initiation. Identification of the SIR in ophiolite complexes implies formation similar
84
to that of forearc lithosphere via seafloor spreading during subduction initiation (Stern and
85
Bloomer, 1992; Stern et al., 2012).
86
The purpose of this paper is not to provide a regional treatment of the tectonic
87
evolution of or new stratigraphic constraints on the early Central American forearc but rather, to
88
examine the early Central American forearc as an extension to cases for which the SIR has been 4
89
tested so far. We do this by focusing on Late Cretaceous and Early Paleogene igneous rocks of
90
Central America and compare their chemotemporal (changes in chemistry with time) progression
91
with the chemotemporal progression of other complexes associated with subduction initiation
92
(forearcs and ophiolites).
93
Crucial to our chemotemporal arguments is recognition that the Sona-Azuero and Golfito proto-
94
arc/arc segments in central Panama and SW Costa Rica, respectively, began to form at least 5
95
M.y. prior to establishment of the Chagres-Bayano Arc in central Panama (see Section 2 for
96
Regional setting and the Supplementary Document for description of these complexes). We
97
document the availability of absolute and relative ages which provides the basis of this
98
recognition in the Supplementary Document (Table S1, see also Table 1).
99 100
2. Regional setting
101
As a result of detailed multidisciplinary studies of Late Cretaceous ‘oceanic igneous
102
terranes’ (terminology of Hauff et al., 1997) and arc-related complexes exposed in the forearc of
103
the Central American Volcanic Arc ( Fig. 1a) in Costa Rica (e.g., Buchs et al., 2010) and Panama
104
(Fig. 1b) (e.g., Hauff et al., 2000a; Tomascak et al., 2000; Geldmacher et al., 2001, 2008; Lissina
105
, 2005; Gazel et al., 2009, 2015; Wörner et al., 2009; Buchs et al., 2009, 2010, 2011; Wegner et
106
al., 2011; Montes et al., 2012a, 2012b; Whattam et al., 2015a, b) we can re-evaluate the
107
chemotemporal evolution of these oldest Central American Volcanic forearc complexes in the
108
context of subduction initiation processes. Most researchers agree that the bulk of the Caribbean
109
Plate is composed of the Caribbean Large Igneous Province (CLIP) (Fig. 1a), an expansive
110
oceanic “plateau” that formed over the Galapagos hotspot prior to its northeasterly drift and
111
insertion between North and South America (see Kerr et al., 2000) prior to subduction initiation.
5
112
Relatedly, the CLIP is interpreted by most as underlying younger arc segments not only in Costa
113
Rica and Panama but also along NW South America in western Ecuador and Colombia all the
114
way to the Leeward Antilles in Curaçao and Aruba (see references in Whattam and Stern,
115
2015a). The ‘standard’ model holds that volcanic arc construction occurred upon the trailing
116
edge of the CLIP in present-day Costa Rica and Panama (Pindell and Kennan, 2009; Wegner et
117
al., 2009; Buchs et al., 2010; Wegner et al., 2011) and elsewhere in the Caribbean realm along
118
the southern margin of the Caribbean Plate in NW South America and the Leeward Antilles. In
119
contrast, on the basis of collated geochemical, isotopic and radioisotopic age datasets of these
120
‘plume-and-arc-related’ complexes in Central America and NW South America which exhibit
121
compositional and temporal overlap between Late Cretaceous hotspot-derived CLIP magmatism
122
and the magmatic effects of subduction initiation, Whattam and Stern (2015a) suggested arc
123
construction in a single, rapidly evolving hybrid plume- and subduction-related environment. By
124
‘plume- and subduction- related’ we mean proto-forearc/forearc complexes that were generated
125
synchronously or nearly so as a result of interaction between the plumehead and the (soon to be)
126
subduction-modified lithosphere that it interacted with; as a result, the complexes which
127
constitute this study were generated from a mantle source that was plume-dominated at first and
128
evolved with time to be modified by subduction. According to this model, arc formation
129
occurred subsequent to subduction initiation which was catalyzed by plume emplacement
130
(plume-induced subduction initiation). A plume-induced subduction initiation mechanism has
131
recently been shown to be valid on the basis of numerical thermomechanical modelling (Gerya et
132
al., 2015).
133
Although the Farallon Plate has been subducting beneath the western Americas since at
134
least 200 Ma, subduction initiation at the Central American Arc was independent of Farallon
6
135
Plate subduction. Central America subduction initiation began between 90-75 Ma along the
136
western periphery of the Caribbean Large Igneous Province (CLIP), which was emplaced within
137
the middle of the Farallon Plate beginning about 92 Ma (Sinton et al., 1997; Kerr et al., 2003 and
138
references therein). Some previous studies of the Central American forearc igneous complexes
139
discussed in this study [see Supplementary Document] suggest that subduction initiation began
140
~75-71 Ma with the magmatic arc growing on pre-existing oceanic plateau (CLIP) basement
141
(Buchs et al. 2010; Wegner et al. 2011]. In contrast, Lissinna et al. (2006) suggest that the infant
142
Panamanian island arc on the western side of the CLIP dates back to 89 Ma. The studies of
143
Lissinna et al. (2006) and Wegner et al. (2011) did not posit a cause for subduction initiation but
144
Buchs et al. (2010) suggested this was tectonically induced, caused by compression of the
145
thickened Caribbean Plate during westward migration of the Americas and collision with South
146
America. As explained above, Whattam and Stern (2015a) in contrast suggested that subduction
147
initiation was the result of plume emplacement and discuss why an induced subduction initiation
148
scenario was unlikely (see Section 5.4; see also Section 5.5 for a discussion of different
149
subduction initiation mechanisms). We note that the timeframe of the Whattam and Stern
150
(2015a) model, which posits subduction initiation as the result of plume emplacement at 90 Ma,
151
is consistent with the conclusions of Lissinna et al. (2006) who suggest subduction initiation by
152
89 Ma.
153
The active Central American volcanic front stretches ~1100 km along the western margin
154
of the Caribbean plate from Costa Rica through Nicaragua, El Salvador and Guatemala to the
155
Guatemala-Mexico border at the southern margin of the North American plate (Figure 1a). The
156
locus of magmatic activity has shifted from east to west after the Eocene. Early plume- and
157
subduction related sequences in the east comprise the ~89-40 Ma Sona-Azuero (Azuero
7
158
Marginal Complex of Buchs et al. 2010), 72-69 Ma Golfito, and 70-39 Ma Chagres-Bayano
159
complexes (Figure 1b, see also Table 1, Table S1 and the Supplementary Document for
160
description of these complexes). The Sona-Azuero and Chagres-Bayano complexes are
161
interpreted as being floored by the 89-85 Ma CLIP (Wegner et al., 2009; Buchs et al., 2010;
162
Wegner et al., 2011; Montes et al., 2012) and the Golfito Complex has been interpreted as both
163
CLIP oceanic plateau (Hauff et al., 2000) and more recently as arc (Buchs et al., 2010). Prior to
164
the study of Buchs et al. (2010), it was generally assumed that most oceanic forearc units along
165
western Costa Rica and Panama represented uplifted segments of the CLIP. Similar
166
interpretations have been made for many oceanic complexes along NW South America and the
167
Leeward Antilles (see Whattam et al., 2015a).
168 169
3. Methods
170
3.1. Petrography
171 172
Fifty new representative thin sections of the Sona-Azuero complex lavas were examined under a
173
petrographic microscope. Sample locations and coordinates are provided in Figure 2 and Table
174
S2, respectively, and selected micrographs comprise Supplementary Figure S1.
175 176
3.2. Elemental analysis
177
Whole-rock samples (n = 25) of the Azuero complex were crushed and ground to powders for
178
major, trace and rare earth element determinations (Table 2). Details of analytical procedures and
179
elemental analysis are provided in the Supplementary Document.
180
8
181
3.3 Data compilation and manipulation
182
3.3.1. Data compilation
183
For comparison with our 25 new whole-rock geochemical analyses of Azuero lavas (Table 2),
184
we also compile geochemical data of lavas of the Sona-Azuero (Buchs et al. 2010; Wegner et al.
185
2011), Golfito (Hauff et al., 2000; Buchs et al., 2010) and Chagres-Bayano complexes (Wegner
186
et al. 2011; Montes et al. 2012b). For further comparison with these Central America forearc
187
complex datasets, we also compile data of global MORB (Gale et al., 2013) and volcanic arc
188
basalts (VAB) of depleted arc systems (PetDB, https://www.earhchem.org/petdb and GEOROC,
189
(http://georoc.mpch-mainz.gwdg.de/georoc/; see Hawkesworth et al., 1997, for definition of
190
‘depleted arc systems’); Late Cretaceous to Eocene ‘plume and arc related’ units (Whattam and
191
Stern, 2015a) in Panama and Costa Rica; lavas from complexes around the circum-Caribbean
192
region interpreted as CLIP; SIR ophiolite lavas (Whattam and Stern, 2011); Izu-Bonin forearc
193
basalts (Reagan et al., 2010; Shervais et al., 2019; and Izu-Bonin ‘forearc-like’ basalts (Arculus
194
et al., 2015)).. The MORB dataset (Gale et al., 2013) is mostly basalt (mean of 50.4 wt. % SiO2)
195
but ranges from 44.9-70.4 wt. % SiO2. Volcanic and plutonic samples comprising Late
196
Cretaceous to Eocene (~85-39 Ma) ‘plume and arc-related’ units (Whattam and Stern 2015a) in
197
Panama and southernmost Costa Rica are considered as either oceanic plateau or proto-arc and
198
arc constructed upon the CLIP (e.g., Buchs et al., 2010) or hybrid plume- and arc-related
199
(Whattam and Stern, 2015a). Basalts interpreted as CLIP (n = 192 tholeiitic and 7 enriched
200
oceanic-island-like basalts) are compiled from GEOROC and are from the Caribbean Sea; the
201
western margin of the Caribbean Plate in Costa Rica (Nicoya, Herradura, Tortugal); the western
202
margin of NW South America in western Ecuador (Piñón Formation and the Pallatanga and
203
Pedernales-Esmeraldas unit) and western Colombia (Gorgona Island, Serrania de Baudo
9
204
Formation and the Western Cordillera); the Lesser Antilles in Aruba (Aruba Lava Formation);
205
and the Caribbean Sea (Colombian and Venezuelan Basins).
206 207
3.3.2. Data manipulation
208
In order to assure that suitable quality analyses of least altered samples of Sona-Azuero, Golfito
209
and Chagres-Bayano were used in our study, only samples which yield oxide totals of 96-102 wt.
210
% (excluding loss on ignition, LOI) are plotted. In our geochemical plots using major elements,
211
oxides are recalculated on an anhydrous (volatile-free) basis and normalized to 100%. However,
212
our geochemical plots and arguments rely primarily on trace elements known to be non-mobile
213
up to greenschist-facies metamorphic conditions.
214 215
4. Results
216
Our lavas are subdivided into three chemical groups on the basis of chondrite-normalized La/Yb
217
concentrations (see section 4.2.1 and Figure 5). Groups I, II and III are analogous to those
218
interpreted by others as plateau basalts, (volumetrically subordinate) enriched basalts and proto-
219
arc or arc basalts (Buchs et al., 2010; Wegner et al., 2011), respectively. As shown by Wegner et
220
al. (2011) and Whattam and Stern (2015a,b), Group I are tholeiitic, Group III range from
221
tholeiitic to calc-alkaline and Group II are analogous to oceanic island tholeiites (Supplementary
222
Figure 2).
223 224
4.1. Petrography
225
Most Group I MORB-like (see Section 5) Azuero lavas are fine-grained tholeiitic basalts.
226
Textures are typically intergranular to intersertal or subophitic and mineralogy is primarily
10
227
plagioclase, clinopyroxene and alteration products (Supplementary Figures S1a, b). In other very
228
fine-grained varieties, clinopyroxene is not obvious and the mineralogy is dominated by thin,
229
wispy, highly altered plagioclase laths. Other samples are coarser-grained and some of these,
230
with large, sericitized plagioclase laths and fresh, subhedral, interstitial clinopyroxene, are
231
texturally similar to Group III arc-like (see Section 5) basalts. The crystallization sequence of
232
plagioclase followed by clinopyroxene is typical of MORB (Bryan, 1983; Pearce et al., 1984).
233
Group II oceanic island basalt-like (see Section 5, Supplementary Figures S1c, d) lavas
234
are subordinate, making up ~10-20 % of the total Azuero samples in our dataset and the
235
combined datasets of Buchs et al. (2010) and Wegner et al. (2011). Overall, these basalts are
236
moderately to heavily altered and distinguished by their chocolate brown coloration and
237
ubiquitous calcite- and chlorite-filled amygdales in the coarser-grained, weakly porphyritic and
238
medium-grained plagioclase phyric samples. Fine-grained varieties of these basalts exhibit
239
micro-porphyritic texture defined by ~1 mm clinopyroxene and 2-5 mm plagioclase phenocrysts.
240
Apatite is an accessory phase (typically <5%).
241
Our Group III arc-like (see Section 5) Azuero lavas are rare (n = 3) and in contrast to
242
lavas of Groups I and II, usually display micro-porphyritic or unique to this group, robustly
243
porphyritic textures defined by large (1.5-2.0 mm), commonly euhedral, fresh clinopyroxene
244
phenocrysts (Supplementary Figures S1e, f). The crystallization sequence, in contrast to the
245
Group I MORB-like lavas, is clinopyroxene followed by plagioclase which is typical of hydrous
246
(i.e., arc) magmas in which early plagioclase crystallization is suppressed (Pearce et al. 1984;
247
Cameron 1985; Sisson and Grove 1994).
248 249
4.2. Major element chemistry
11
250
4.2.1. MgO vs. TiO2, Al2O3, FeOt and K2O
251
As developed below in Section 4.3, the Sona-Azuero lavas of Buchs et al. (2010), Wegner et al.
252
(2011) and our dataset are subdivided into three groups on the basis of chondrite-normalized
253
La/Yb concentrations. Group I are interpreted by Buchs et al. (2010) and Wegner et al. (2011) as
254
plateau basalts; Group II as enriched basalts (Buchs et al., 2010); and Group III as proto-arc or
255
arc volcanics (Buchs et al., 2010; Wegner et al., 2011). These groupings are referred to below.
256
A plot of SiO2 vs. MgO (Figure 3) shows that lavas considered in this study range from
257
basaltic to rhyolitic with MgO concentrations of 1.06 to 11.68 wt.%. Fractionation patterns for
258
MORB (n = 14, 782) and depleted (see Hawkesworth et al., 1997) arc systems (n = 4, 379) are
259
broadly similar but with the latter ranging to lower MgO at low SiO2 contents (˂55 wt.%) and
260
higher MgO at higher SiO2 (55-72 wt.%). Despite the considerable overlap of MORB and VAB,
261
a large number of Sona-Azuero Group III arc (n = 8) and Chagre-Bayano samples (n = 7) exhibit
262
higher arc-like MgO than MORB at silica concentrations ˃ 55 wt.% SiO2. As well, two Sona-
263
Azuero proto-arc, three Sona-Azuero Group II samples (including two from this study) and one
264
Chagres-Bayano sample display lower SiO2 than MORB for a given MgO concentration at low
265
SiO2 (45-48 wt. %) and fall within the arc domain.
266
In Figures 4a-d, we plot MgO vs. TiO2, Al2O3, FeOt and K2O for basalts of our Azuero
267
dataset alongside basalts of the Sona-Azuero, Golfito and Chagres-Bayano suites. For
268
comparison, also plotted on Figure 4 are lavas of MOR and depleted arc systems (basalts only).
269
In all plots there exists considerable overlap between MORB and VAB but some samples plot
270
uniquely within either field. In consideration of MgO vs. TiO2, all six Sona-Azuero Type II
271
basalts plot uniquely within or just above the MORB field and outside of the arc field. Two of
12
272
our three Type II Sona-Azuero basalts also have Al2O3 and FeOt concentrations which plot
273
within the MORB field. In MgO vs. TiO2 and the remainder of the MgO vs. major oxides plots
274
(Al2O3, FeOt and K2O), between two to four low-magnesium (~3-6 wt.% MgO) Chagre-Bayano
275
lavas plot uniquely within the arc field. Similarly, apart from K2O, one low MgO (~4.2 wt.%
276
Mg) Sona-Azuero proto-arc sample plots distinctively within the arc field; two of our Sona-
277
Azuero Type III arc lavas have TiO2 and FeOt akin to arc basalts. Additionally, two high MgO
278
(~10-12 wt. %) Chagre-Bayano lavas exhibit MgO vs. TiO2, Al2O3, FeOt and K2O compositions
279
that plot distinctly within the arc field or on the cusp between MORB and arc. The remainder of
280
samples plot within the overlapping area between MORB and VAB.
281
4.3. Trace element chemistry
282
4.3.1. MgO vs. Ce, Yb, Y and Zr
283
We also plot MgO vs. a light and heavy rare earth element (LREE, Ce and HREE, Yb) and two
284
high field strength elements (Y, Zr) for basalts of our Azuero dataset alongside basalts of the
285
Sona-Azuero, Golfito and Chagres-Bayano suites for comparison with MORB and depleted arc
286
system basalts (Figures 4e-h). In terms of MgO vs. Ce, two of our Sona-Azuero high-TiO2 Group
287
II samples plot above both the MORB and arc fields with the highest Ce of all samples (46-47
288
ppm) whereas two Sona-Azuero Group II samples of Buchs et al. (2009) plot within the MORB
289
field along with one Sona-Azuero proto-arc and arc sample. Many low-magnesium (˂6 wt.%
290
MgO) basalts plot within the arc field including two of our Sona-Azuero Group III arc lavas,
291
three Sona-Azuero Group I lavas interpreted as plateau by Buchs et al. (2009) and Wegner et al.
292
(2011), three Sona-Azuero Group III arc basalts and five Chagres-Bayano basalts. On the other
293
end of the MgO spectrum, one high-MgO (~11 wt. % MgO) Group I Sona-Azuero and one
13
294
Chagres-Bayano basalt (~12 wt. % MgO) also plot within the arc field. The remaining basalts
295
plot where the MORB and arc fields overlap in MgO vs. Ce space.
296
On a plot of MgO vs. Yb (Fig. 4f), four Sona-Azuero Group III arc basalts including all
297
three of our Sona-Azuero Group III basalts plot alongside six Chagres-Bayano basalts uniquely
298
within the low-HREE arc field. Also, three Sona-Azuero Group I, six Sona-Azuero Group III
299
proto-arc and arc and six Chagre-Bayano basalts plot outside of both the MORB and arc field;
300
the remaining majority of samples plot within the MORB realm with Yb ˃ ~1.4.
301
Many samples exhibit arc-like HFSE concentrations. On a MgO vs. Y plot (Fig. 4g), our
302
three Sona-Azuero Group III basalts, four of six Sona-Azuero Group II basalts and eight Sona-
303
Azuero Group III proto-arc and arc basalts fall alongside almost half of the Chagre-Bayano
304
basalts (12 of 26 samples) within the arc field. Our three Sona-Azuero Group III basalts and
305
Chagres-Bayano basalts exhibit the lowest Y concentrations (˂14 ppm). The remaining basalts
306
fall within an area of MORB and arc overlap. A similar trend is seen on MgO vs. Zr (Fig. 4h)
307
with our Sona-Azuero Group III and Chagres-Bayano basalts exhibiting the lowest Zr arc-like
308
contents of all samples.
309
4.3.2. Subdivision of Sona-Azuero lavas into three groups on the basis of [La/Yb] CN
310
Table 3 provides the foundation for subdivision of our Azuero lavas into three groups on the
311
basis of chondrite-normalized La/Yb concentrations ([La/Yb]CN).
312
Compositionally, our Azuero Group I basalts are slightly to moderately light rare earth
313
element depleted ([La/Yb]CN = 0.75-1.09, mean = 0.96) and compositionally analogous to the
314
Group I (non-enriched) Azuero ‘plateau’ samples of Buchs et al. (2010) and the ‘CLIP’ basalts
315
of Wegner et al. (2011) ([La/Yb]CN = 0.63-1.15, mean = 0.92, Table 3).
14
316
Our Group II Azuero basalts are volumetrically subordinate, light rare earth element-
317
enriched ([La/Yb]CN = 2.29-2.86) and compositionally similar to the Group II ‘enriched plateau’
318
samples of Buchs et al. (2010) ([La/Yb]CN = 1.92-2.81) (Table 3). Taking all enriched samples
319
into consideration (i.e., of Buchs et al., 2010 and ours), these basalts are perhaps best described
320
as oceanic island tholeiites (OIT on Supplementary Figure 2) on the basis of TiO2 vs. Y/Nb
321
relations (Floyd and Winchester, 1975; Winchester and Floyd, 1977).
322
On the basis of chondrite-normalized rare earth element and N-MORB-normalized plots
323
(see below), our Group III lavas are VAB-like and analogous to the Azuero Proto-Arc and Arc
324
samples of Buchs et al. (2010) and the ‘CLIP’ Arc samples of Wegner et al. (2011), although our
325
samples range to lower absolute rare earth element concentrations. Rare earth element patterns of
326
Group III lavas range from slightly light rare earth element-enriched ([La/Yb]CN = 1.28-1.58],
327
Table 3) to flat to slightly ‘U-shaped’ (see next section), i.e., middle rare element-depleted, e.g.,
328
two samples have [La/Gd]CN > 1 and [Yb/Gd]CN > 1).
329 330
4.3.3. Chondrite-normalized rare earth element and N-MORB-normalized incompatible element
331
plots
332 333
Chondrite-normalized rare earth element and N-MORB-normalized plots of Sona-Azuero,
334
Golfito and Chagres-Bayano basalts are provided in Figs. 5 and 6, respectively.
335
Although the rare earth element patterns of Golfito and Chagres-Bayano basalts are
336
similar to those of Sona-Azuero there are some significant differences. Rare earth element
337
patterns of proto-arc lavas of the Golfito Complex (Fig. 5d) generally fall within the range of the
338
Group I and Group III Sona-Azuero lavas interpreted as oceanic plateau and proto-arc/arc
15
339
respectively, but range to very low absolute rare earth element abundances (mean ∑ rare earth
340
element of 32) and light rare earth element fractionations (e.g., [La/Yb]CN of 0.81). These values
341
are higher only than that of the Chagres-Bayano Arc (mean ∑ rare earth element ~30) and the
342
lone Chagres-Bayano basalt interpreted as oceanic plateau ([La/Yb]CN of 0.47) (Figs. 5e, f).
343
Relatedly, this [La/Yb]CN ratio of the lone Chagres-Bayano basalt interpreted as oceanic plateau
344
(0.47) is much lower than that of the mean of Sona-Azuero oceanic plateau and arc basalts (with
345
mean [La/Yb]CN of 0.93 and 1.54, respectively). Although the Chagres-Bayano Arc exhibits
346
mean [La/Yb]CN intermediate to that of the Sona-Azuero oceanic plateau and proto-arc/arc (mean
347
[La/Yb]CN of 1.18 vs. 0.93 and 1.54, respectively), the absolute rare earth element abundances of
348
Chagres-Bayano Arc lavas range to much lower concentrations (mean ∑ rare earth elements of
349
30) of ~ 3x chondrite (Fig. 5f).
350
In terms of N-MORB-normalized incompatible element patterns, the most conspicuous
351
differences between our Azuero Groups are: the lack of large ion lithophile element enrichment
352
in fluid-mobile elements (e.g, Rb, Ba, K, Pb) in Group II, which is well developed in both
353
Groups I and III; the lack of a negative Nb-anomaly (with respect to Th and Ce) in Groups I and
354
II, but which is obvious in Group III; and the pattern of Y and the heavy rare earth elements (Yb,
355
Lu), which are flat in Groups I and III but which decrease with decreasing incompatibility in
356
Group II (Figs. 6a-c). The differences between the N-MORB-normalized patterns of the Sona-
357
Azuero basalts interpreted as CLIP plus our Group I basalts interpreted as segments of the
358
earliest hybrid, plume-dominated unit and basalts interpreted as arc are modest, apart from the
359
prominent Nb depletion and the slightly more, fluid-modified compositions of the latter. For
360
example, although the CLIP basalts exhibit no obvious negative Nb-depletion, they do exhibit
361
modest negative Zr and Ti anomalies and contain only slightly less Ba and Pb (Fig. 6a). 16
362
N-MORB normalized plots of Golfito and Chagres-Bayano basalts are provided in Figs.
363
6d-f. Golfito proto-arc basalts fall completely within the range of Sona-Azuero oceanic plateau
364
and proto-arc/arc basalts but range to lower concentrations of large ion lithophile elements (Ba,
365
K, Pb and Sr) (Fig. 6d). Although a negative Nb anomaly is not well developed in the lone
366
Golfito sample interpreted as oceanic plateau (Hauff et al., 2000) it does exhibit a subtle negative
367
Zr anomaly and concentrations of U, Pb, K and Sr higher than or similar to Golfito basalts
368
interpreted as arc (Buchs et al., 2010); the Golfito proto-arc samples of Buchs et al. (2010)
369
exhibit subtle but clear negative Nb anomalies (Fig. 6d)). The N-MORB-normalized
370
incompatible element plot of the lone Chagres-Bayano sample interpreted as oceanic plateau
371
(Fig. 6e) is somewhat anomalous as it lacks the high LILE enrichments exhibited by most other
372
arc (e.g., Chagres-Bayano Arc, Fig. 6f) and oceanic plateau samples.
373 374
5. Discussion
375
In the following sections, we explore the significance of our new geochemical data and that of
376
other workers to address the following questions: (1) Do Late Cretaceous – Paleogene igneous
377
rocks of Panama and Costa Rica follow the “subduction initiation rule” promulgated for the Izu-
378
Bonin-Mariana arc and many ophiolites?; (2) What is the nature of the contact between “oceanic
379
plateau” and arc and chemotemporal evolution of magmatism?; and (3) What was the nature of
380
the early arc system?
381 382
5.1 Evidence of adherence of the early Central American Volcanic Arc system to the
383
subduction initiation rule
17
384
We focus here on elements that are not mobilized by hydrothermal alteration and metamorphism
385
for the purpose of constraining mantle source fertility, degrees of partial melting and mantle
386
source fugacity. Ratios of Zr/Y, Nb/Y (and Nb/Yb) are useful as these decrease with mantle
387
depletion and as partial melting increases. Because of the compatibility of Y and Yb in garnet,
388
these ratios can also reflect if garnet was present during melt generation but this may be the case
389
only in Group II enriched oceanic island tholeiite-like basalts which exhibit low Y and heavy
390
rare earth element relative to middle rare earth element ( Fig. 5b). In Figures 7-9, in addition to
391
plotting Panamanian and Costa Rican samples from this study, we also plot SIR ophiolite basalts
392
and basaltic andesites for comparison. We do this because SIR ophiolite lavas are (i) thought to
393
have formed during SI in a proto-forearc environment and show a distinctive chemotemporal
394
(change in chemistry with time) trend from less- to more-depleted and subduction–modified
395
compositions with time (i.e., MORB- to VAB-like) (Whattam and Stern, 2011) which (ii) along
396
with other chemical criteria (see below), are critical for understanding the evolution of the
397
mantle sources for Sona-Azuero and Chagres-Bayano.
398 399
5.1.1. Source fertility
400
In Fig. 7, we plot Zr vs Zr/Y. As demonstrated by Whattam and Stern (2011) and shown in Fig.
401
7a, whereas the older lavas of the earliest-formed lower subduction initiation rule ophiolite units
402
are mostly MORB-like (84% of the lower unit basalts plot as MORB), those of the younger,
403
uppermost unit in contrast plot completely within the VAB field.
404
These relations show that forearc units progress from tapping/melting of a less to more
405
depleted source or lower to higher degrees of partial melting over the course of seafloor
406
spreading during subduction initiation. Notably, all tholeiiitic Group I Sona-Azuero basalts
18
407
interpreted as CLIP (Fig. 7b) plot within the compositional field defined by all (Group III) Sona-
408
Azuero and most Chagres-Bayano and Golfito basalts interpreted as proto-arc (Fig. 7c)
409
suggesting a similar source or degrees of partial melting for Sona-Azuero oceanic plateau and
410
arc. As well, Fig. 7c confirms what was shown by the N-MORB and rare earth element plots,
411
i.e., that Chagres-Bayano arc basalts were derived from a more depleted source, or one that has
412
undergone higher degrees of partial melting than Sona-Azuero. Similarly, the Golfito basalts
413
exhibit evidence of derivation from a source intermediate in composition to that of Sona-Azuero
414
and Chagres-Bayano and Sona-Azuero Group III basalts are more similar in composition to
415
Sona-Azuero Group I than to Chagres-Bayano. Geochemical relations demonstrate a MORB to
416
VAB transition from the Sona-Azuero to the Chagres-Bayano unit and demonstrate adherence to
417
the SIR for the early Central American Volcanic forearc.
418 419
5.1.2. Degrees of partial melting
420
Fig. 7 shows that the source(s) of the Chagres-Bayano basalts and our Group III Azuero basalts
421
were either more depleted or underwent higher degrees of partial melting than the Sona-Azuero
422
source. These plots however, don’t distinguish between the two possibilities, and if the latter is
423
accurate, provide estimates of the relative degrees of partial melting.
424
Cr vs. Y (Fig. 8) can be used to determine relative degrees of partial melting of basalts
425
(Murton 1989). Superimposed on Fig. 8 are sources 1, 2 and 3 (S1, S2, S3) which represent a
426
MORB source, a more depleted source after 20% MORB extraction, and a most depleted source
427
after ~12% extraction of S2, respectively. We caution that the F values discussed below are
428
model dependent and we are interested not in absolute degrees of partial melting but differences
429
in partial melting between lavas of Sona-Azuero and Chagres-Bayano. Also shown in Fig. 8a are
19
430
fields defined by MORB, VAB and boninite (Pearce, 1982). Inspection of Fig. 4b reveals that
431
whereas the lower basalts of subduction initation rule ophiolites are consistent with ~20-40%
432
partial melting of S1 or 20% partial melting of S1 and ~10% partial melting of S2, the upper unit
433
basaltic andesites of subduction initation rule ophiolites indicate higher degrees of partial melting
434
(~10-40% partial melting of S2). A similar relation is seen for Sona-Azuero vs. Chagres-Bayano
435
lavas (Figs. 8c, d). Most Sona-Azuero basalts interpreted as oceanic plateau plot within the
436
MORB field and are consistent with ~20-40% partial melting of S1 (Fig. 8c). Approximately half
437
of the Sona-Azuero basalts interpreted as arc and some Chagres-Bayano arc basalts also plot
438
within the MORB field consistent with ~ up to 40% partial melting of S1, or perhaps more
439
realistically~ 12% melting of S2 (Fig. 8d). The other half of the Sona-Azuero basalts interpreted
440
as arc along with the Golfito basalts record partial melting estimates of 10-20% of S2. Compared
441
to SI ophiolites, the Chagres-Bayano Arc basalts clearly exhibit evidence of derivation from the
442
highest degrees of partial melting. About 35% of the Chagres-Bayano arc basalts plot within the
443
boninite field and are consistent with derivation via up to about 16-17% partial melting of S3.
444
Although boninites require >52 wt.% SiO2, >8 wt.% MgO and <0.5 wt.% TiO2 according to the
445
IUGS (LeBas, 2000), some of these Chagres-Bayano basalts which plot as boninite contain >8
446
wt.% MgO and <0.5 wt.% TiO2 (e.g., samples P-010223-6, PAN-03-016, Wegner et al., 2011).
447
Such high degrees of partial melting reflect exceptionally hot and shallow forearc mantle beneath
448
young intra-oceanic subduction-related systems (e.g., Tatsumi and Eggins, 1995). Like the Zr vs
449
Zr/Y plot (Fig. 7) the Cr/Y plot (Fig. 8) demonstrates an unequivocal chemotemporal evolution
450
that follows the SIR.
451 452
5.1.3. Source oxygen fugacity
20
453
It has long been recognized that MORB are distinguishable from VAB on the basis of Ti/V due
454
to the more oxidized nature (lower Ti/V ratio) of arc environments vs. the more reduced nature
455
of MORB sources (Shervais, 1982). We plot SIR ophiolite lavas vs. the older and younger
456
segments of the Sona-Azuero, Chagres-Bayano and Golfito complexes in Ti/V space in Figs. 9a-
457
c to demonstrate that the features developed above in various incompatible and immobile
458
elements hold true when considering source fugacity; i.e., whereas the older Sona-Azuero suite
459
(basalts interpreted as oceanic plateau and proto-arc/arc) is consistent with derivation from a
460
MORB-like source, basalts from Chagres-Bayano on the other hand are clearly more consistent
461
with derivation from a more arc-like source.
462
In Figs. 9d-f, we also plot FAB from the Izu-Bonin Mariana Arc system (Reagan et al.,
463
2010), Exp. 352 forearc basalts from the outer Izu-Bonin Mariana Arc system (Shervais et al.,
464
2019), and Exp. 351 Site U1438 (Arculus et al., 2015) forearc basalt-like lavas (which presently
465
occupy the rear arc of the Izu-Bonin Mariana Arc system). Lowermost basalts of the Izu-Bonin
466
Mariana forearc were first interpreted as trapped Phillipine Sea Plate (deBari et al., 1999), a
467
similar interpretation of which was applied to the Tonga forearc (Meffre et al., 2012). However,
468
these first-formed lavas (forearc basalt) are MORB-like and instead mark the first magmatic
469
expression of subduction initiation (Reagan et al., 2010). As shown in Fig. 9e, the vast majority
470
of the older MORB-like Sona-Azuero and the lone Chagres-Bayano basalt (all considered as
471
oceanic plateau) and Golfito basalts plot within the fields defined by the forearc basalts.
472
Similarly, the vast majority of the younger Chagres-Bayano Arc basalts also plot within the field
473
defined by forearc basalts (Fig. 9f), whereby most (77%) of the older Sona-Azuero basalts
474
interpreted as arc plot outside the fields defined by Izu-Bonin Mariana (Ishizuka et al., 2006;
475
Reagan et al., 2010) and Exp. 352 forearc basalts (Expedition 352 Scientists, Preliminary Report
21
476
2015) within the MORB field. This again demonstrates a SIR chemotemporal evolution from
477
early MORB-like lavas to later arc-like lavas, this time on the basis of oxygen fugacity, in the
478
early Central American forearc.
479 480
5.2. Isotopic constraints on the plume-contaminated nature of the source
481
A myriad of studies show the plume-contaminated nature of Central American forearc lavas as
482
the result of emplacement of the CLIP which immediately preceded and instigated subduction
483
initiation (plume-induced subduction initiation, Whattam and Stern (2015a) and references
484
therein). In terms of source components and source mixing, it was shown by Allibon et al.
485
(2008) that the isotopic compositions of Ecuadorian arc or arc-like lava sequences (which
486
according to Whattam and Stern 2015a are essentially analogous to arc lavas in Sona-Azuero,
487
Golfito and Chagres-Bayano) built upon CLIP oceanic plateau fragments could be explained in
488
terms of tripolar mixing of three components: a dominant depleted MORB mantle component; a
489
high µ (where µ is a 238U/204Pb ratio of an Earth reservoir) component carried by the CLIP; and a
490
subducted pelagic sediment component. It was further shown by Allibon et al. (2008) that an
491
average contribution of 70-75% depleted MORB mantle, 12-15% high µ , and 7-15% pelagic
492
sediments (cherts) is required to explain their composition. Furthermore, Whattam ( 2018)
493
showed that the Central American lavas in this study also plot within the
494
206
495
Cretaceous Ecuadorian lavas in the study of Allibon et al. (2008). It is for these reasons that we
496
use the term ‘plume contaminated’ to describe the forearc basalts of Sona-Azuero and the VAB
497
of Chagres-Bayano.
143
Nd/144Nd vs
Pb/204Pb field defined by the Galapagos Plume at 90 Ma similarly to the western Late
498
22
499
5.3. Nature of the contact between “oceanic plateau” and arc and chemotemporal evolution of
500
magmatism
501 502
Buchs et al. (2010) present a synthetic ‘tectonostratigraphic’ chart (their Fig. 3, reproduced in
503
part in our Fig. 10a) but there is a ~10-12 M.y. hiatus between inferred termination of oceanic
504
plateau construction at 85 Ma and incipient arc construction at 75-73 Ma (timing of SI as
505
interpreted by Buchs et al., 2010) at both the Azuero marginal and Golfito complexes. The only
506
study we know of that documents the nature of the contact between the lower “oceanic plateau
507
unit” and the upper arc unit in the Sona-Azuero Complex is that of Corral et al. (2011) (their Fig.
508
2, reproduced in our Fig. 10b). They (Corral et al. 2011) show a conformable contact between
509
the oceanic plateau basement and the overlying proto-arc sequence, which in turn is conformably
510
overlain by the arc sequence (the Rio Quema Formation, see Fig. 10b). This bolsters the
511
supposition of Whattam and Stern (2015a) of a continuous, uninterrupted magmatic succession
512
between underlying lavas interpreted as plateau and overlying ones interpreted as proto-arc/arc.
513
Lissinna et al. (2006) report that earliest arc activity commenced at 88.3±5 Ma but unfortunately,
514
no further details are available. If earliest formation of the Panamanian Arc was indeed ~90-89
515
Ma, this supports the model of Whattam and Stern (2015a) that plateau and arc magmatism
516
overlapped beginning about 90 Ma. We also note that Buchs et al. (2010) remark that some
517
earliest supra-subduction zone proto- igneous rocks are indistinguishable from those of ‘typical’
518
oceanic plateaus which also hints that plateau and arc magmatism overlapped (or that earliest SI
519
lavas tapped a strongly plume-contaminated source). Corral et al. (2011) provide a stratigraphic
520
section though the Rio Quema Formation (Azuero), showing that the Azuero proto-arc group
521
simply overlies the Azuero (oceanic plateau) igneous basement which, if the ages of Lissinna et
23
522
al. (2006) are accurate, suggest a continuous, uninterrupted interval from oceanic plateau to
523
proto-arc/arc magmatism. This, coupled with the complete compositional overlap between Sona-
524
Azuero lavas and intrusives interpreted as oceanic plateau and those interpreted as proto-arc and
525
arc suggest that both are related in space and time. Clearly, more stratigraphic and
526
geochronologic investigations are needed but the data in hand support a model whereby the
527
evolution from earlier CLIP-like basalts to younger arc-like lavas was continuous and
528
uninterrupted (Fig. 11a) as opposed to a scenario whereby arc construction occurred upon a pre-
529
existing plateau (Fig. 11b).
530
It is increasingly clear that the subdivision between oceanic plateau and arc magmatic
531
sequences in Central America are gradational. We recognize that the subdivision is based on
532
lavas and intrusives with negative primitive-mantle normalized high-field strength element
533
anomalies (particularly Nb) being assigned as arc and those without as oceanic plateau.
534
However, it was shown by Whattam and Stern (2011) that the first magmatic expression of some
535
subduction initation systems is MORB-like, and some basalts of lower unit subduction initiation
536
ophiolites exhibit weak primitive mantle normalized negative Nb-anomalies, others do not; a
537
robust negative Nb-anomaly and associated large ion lithophile enrichment is typically seen
538
only in younger lavas and intrusives. We agree with the interpretation of Whattam and Stern
539
(2015a) that the earliest magmatic activity at the Central American Arc system in Panama and
540
Costa Rica was the result of tapping of a hybrid, plume-subduction source as the result of plume-
541
induced subduction initiation. As shown in Sections 5.1.1.-5.1.3., whereas lavas of both Sona-
542
Azuero units - i.e., those interpreted as oceanic plateau and those interpreted as arc - are
543
‘MORB-like’, the lavas of the Chagres-Bayano unit are mostly ‘arc-like’. We conclude on this
24
544
basis that magmatism evolved continuously from plume-contaminated MORB- to VAB-like with
545
time consistent with the SIR.
546 547
5.4. Subduction initiation mechanisms
548
As explained by Stern (2004) and later by Stern and Gerya (2018), subduction initiation can be
549
either spontaneous or induced. In the case of the former, there are three possible types: transform
550
collapse, passive margin collapse, and plume head margin collapse (i.e., plume-induced
551
subduction initiation). In contrast, induced subduction initiation is triggered by continued plate
552
convergence following collision, for example of an oceanic plateau with a subduction zone.
553
There are two types of induced subduction initiation: a new subduction zone of similar dip to the
554
original subduction zone can initiate behind the collision (transference) or in front of the
555
collision and (polarity reversal).
556
The earliest workers to propose spontaneous subduction initiation at a lithospheric
557
weakness (transform separating older, denser lithosphere from younger, warmer lithosphere) was
558
Matsumoto and Tomoda (1983). This idea of spontaneous subduction initiation along a
559
transform fault was adopted by Stern and Bloomer (1992) to explain the early evolution of the
560
Izu Bonin Marianas arc system and by others for other regions of forearc formation (e.g.,
561
Whattam et al., 2006, 2008; Shafaii Moghadam et al., 2013; Zhou et al., 2018). Other studies
562
have focused on forced convergence across fracture zones (e.g., Hall et al., 2003). Subduction
563
nucleation along a passive margin (Marques et al., 2014; Ueda et al., 2008; Burov and Cloetingh,
564
2010; Van der Lee et al., 2008) is implied in the Wilson (1966) cycle but until very recently
565
(Pandey et al., 2019), there are no known Cenozoic examples (Stern, 2004). Pandey et al. (2019)
25
566
recovered forearc basalts and overlying boninitic-like forearc basalts in the Laxmi Basin, western
567
Indian passive margin, which record geochemical and isotopic attributes identical to first formed
568
lavas of the Izu Bonin Marianas and supra-subduction zone ophiolites and may provide the first
569
evidence for subduction initiation along a passive margin. Perhaps the best example of a induced
570
subduction initiation polarity reversal event is that caused by collision of the Ontong Java
571
Plateau, the world’s largest igneous province, which attempted to subduct to the SW at the Vitiaz
572
Trench ~10-4 Ma (Phinney et al., 1999; Cooper and Taylor, 1985). Thick, buoyant lithosphere of
573
the Plateau jammed the subduction zone and caused a subduction polarity flip to the NE. As
574
explained in the Introduction, Buchs et al. (2010) suggested that subduction initiation at the
575
Central American Arc was tectonically induced and resulted from compression of the thickened
576
Caribbean Plate during westward migration of the Americas and collision of the thickened
577
Caribbean Plate with South America. As explained above in Section 5.3, on the basis of a
578
comprehensive set of geochemical, isotopic and geochronologic data and tectonic considerations,
579
we concur with Whattam and Stern (2015) that arc construction in Central America began
580
instead about 90 Ma as the result of plume-induced subduction initiation. A second example of
581
plume-induced subduction initiation has been recently proposed for the Cascadia subduction
582
zone in the Pacific NW, USA (Stern and Dumitru, 2019) and plume-induced subduction
583
initiation has shown to be viable on the basis of numerical thermomechanical modelling (Gerya
584
et al., 2015).
585 586
5.5. Physical nature of the early arc system and tectonic considerations
587
In their study of Panamanian magmatism circa 75 Ma to present, Wegner et al. (2011) suggest
588
that arc magmatism began in the Sona and Azuero peninsulas at ~71 Ma (Sona-Azuero Arc), 26
589
then shifted to the Chagres-Bayano region at 66 Ma; Wegner et al. (2011) also apparently
590
consider these as two distinct arcs. We disagree with these notions for several reasons. Firstly,
591
arc lengths are typically not at the 100s of km scale but rather 1000s of km scale. For example,
592
the modern-day Central American Volcanic Arc system is ~1500 km long and the Izu-Bonin
593
Marianas Arc system is ~2800 km long. The magmatic arc ‘belts’ along the Sona-Azuero
594
peninsula and the Chagres-Bayano region are only ~200-250 km long, but restoration of
595
paleomagnetically derived vertical-axis rotations (Montes et al., 2012a) suggests that the
596
combined length of the Campanian to Eocene Central American arc was at least 1000 km.
597
Furthermore, combined geochemical and geochronological evidence suggests that the nascent
598
Central American arc stretched westwards from the Nicoya peninsula in western Costa Rica
599
some 200 km south of Nicaragua, to either present-day western Ecuador or Colombia some 1400
600
km to the east (Whattam and Stern, 2015a). Secondly, the age constraints of Lissinna (2005)
601
indicate that Sona-Azuero magmatic activity was contemporaneous with Chagres-Bayano
602
magmatism between 70-40 Ma. Thirdly, unequivocal ‘arc’ magmatism at the Sona-Azuero and
603
Golfito complexes began by at least 75 Ma (Buchs et al., 2010) but the 89-85 Ma Sona-Azuero
604
CLIP MORB-like oceanic plateau basalts are largely indistinguishable from the younger Sona-
605
Azuero arc and proto-arc basalts. Furthermore, Lissinna et al. (2006) suggest arc magmatism
606
began by ~90 Ma. Finally, as shown by Whattam and Stern (2015a), most (~90-85 Ma) oceanic
607
plateau basalts from the western edge of the CLIP (Costa Rica, e.g., Nicoya, Herradura,
608
Tortugal) including those of Sona-Azuero and the NW segment of the South American Plate in
609
Ecuador and Western Colombia, exhibit subduction modified compositions. Although a lateral
610
west to east transition in chemistry cannot be categorically ruled out, the relations in the
611
chemistry of Sona-Azuero vs. Chagres-Bayano basalts are more consistent with a
27
612
vertical/stratigraphic transition, although eruptions may have become more arc-like as
613
magmatism retreated from the trench. This chemostratigraphic progression can also be
614
considered as chemotemporal as the change in magmatic affinities is largely vertical and hence a
615
record also of change in chemistry with time; this chemotemporality is identical to that recorded
616
by the slightly older Sona-Azuero and the younger Chagres-Bayano complexes. Thus, we
617
suggest that the Chagres-Bayano arc segment represents a vertical as well as lateral migration of
618
arc magmatism that is also seen for Sona-Azuero circa 71 Ma (Wegner et al., 2011) (Figure 7).
619
In other words, the Chagres-Bayano arc segment may represent the carapace of the early
620
established Central American Arc system.
621
If Chagres-Bayano does indeed represent the latest phase of early Central American Arc
622
magmatism, the present position of the Chagres-Bayano arc segment indicates that it is offset to
623
the north. In the tectonic model of Whattam et al. (2012), the early (pre-40 Ma) Central
624
American Arc was shut down at about 40 Ma as the result of collision of an oceanic plateau with
625
the arc system (see also Kerr and Tarney, 2005). As explained by Montes et al. (2012) and
626
Whattam et al. (2012), this collision caused left-lateral offset of the originally contiguous early
627
(pre-40 Ma) Central American Volcanic Arc system which stretched from at least western Costa
628
Rica in the NW to western Ecuador or Colombia in the SE (Whattam and Stern, 2015a) and
629
resulted in a >100 km displacement to the north between 40-30 Ma (Lissinna, 2005; Montes et
630
al., 2012). Alternatively, left-lateral offset may have been caused by a shift in the subducted
631
plateau from Azuero to a region to the east (Lissinna et al., 2006) or as the result of collision of
632
South America with the Panamanian Arc system (Farris et al., 2011; Barat et al., 2014).
633
Regardless of the mechanism of displacement or the nature of the colliding body, we suggest that
28
634
the Chagres-Bayano arc segment was translated to the north after originally forming the carapace
635
and rear of the early arc system subsequent to slab rollback but prior to displacement (Fig. 11).
636 637
6. Conclusions
638 639
An unambiguous chemotemporal transition from early-formed plume-contaminated MORB-like
640
to VAB-like lavas is seen in Late Cretaceous-Paleogene magmatic arc sequences in Panama as
641
revealed by incompatible trace element compositions which illustrate fundamental differences
642
between the older Sona-Azuero complex and the younger Chagres-Bayano complex. While the
643
trace element signatures of basalts from both units are similar, the absolute REE concentrations
644
of Chagres-Bayano basalts range to much lower Σ rare earth element concentrations (12-52 ppm)
645
than those of Sona-Azuero (Σ rare earth elements = 33-102 ppm), exhibit more negative high
646
field strength element-anomalies and higher fluid mobile element/immobile element ratios.
647
Relative to Sona-Azuero, the 70-39 Ma Chagres-Bayano basalts reflect higher degrees of partial
648
melting of a more depleted and oxidized mantle source that received greater slab-derived shallow
649
and deep subduction contributions.
650
Moreover, the Sona-Azuero basalts interpreted as proto-arc and arc are compositionally
651
more similar to the Sona-Azuero basalts interpreted as oceanic plateau than they are to the
652
Chagres-Bayano basalts, which if age estimates are accurate suggest that CAVAS subduction
653
began by ~89 Ma. The overarching chemotemporal trend recorded by Sona-Azuero and Chagres-
654
Bayano sequences may reflect the combined effects of a west to east migration of magmatism at
655
68 Ma as well as a vertical/stratigraphic transition. We suggest that the arc segment now
656
preserved at Chagres-Bayano was originally generated above and to the rear of Sona-Azuero
29
657
forearc magmatism and was displaced to the north perhaps due to collision with an oceanic
658
plateau on the Cocos plate, which shut down the early arc system ~40 Ma.
659
The chemotemporal progression from older MORB-like lavas and intrusives at Sona-
660
Azuero to younger VAB-like lavas at Chagres-Bayano to the east confirms adherence of the Late
661
Cretaceous to Eocene Central American forearc to the SIR, similar to Tethyan ophiolites and the
662
Izu-Bonin-Mariana forearc. Importantly, we note that this adherence is independent of the two
663
competing tectonic models described. In other words, even if arc construction occurred intrusive
664
into and atop a pre-existing oceanic plateau and we exclude the MORB-like affinities of the
665
oceanic plateau, the SIR still applies as the Sona-Azuero Proto-Arc and arc lavas and intrusives
666
are also MORB-like in contrast to those of the Chagres-Bayano Arc which instead exhibit
667
unequivocal volcanic arc affinities.
668 669
Acknowledgements
670 671
We are grateful to Mark Tupper, NSF EAR-0824299, National Geographic, Smithsonian
672
Institution, and Ricardo Perez S.A. which partially funded this project. S. Zapata and D. Ramirez
673
are acknowledged for their help in the field and for generating the base map with the sample
674
localities. Access to field areas and collection permits were granted by Ministerio de Industria y
675
Comercio of Panamá. We thank Jörg Geldmacher, Jeffrey Ryan, and Yamirka Rojas-Agramonte
676
for their reviews of a previous version of the manuscript and Fernando Marques, Kosuke Ueda
677
and an anonymous reviewer for their reviews of the current manuscript and positive comments
678
which resulted in a greatly improved paper.
679
Figure Captions
30
680
Figure 1 (a) Present tectonic configuration of the Circum-Caribbean region. (b) Magnification
681
and detail of the boxed region in Figure 1a showing the distribution of Late Cretaceous to middle
682
Eocene (~75-39 Ma) arc magmatism in Golfito, easternmost Costa Rica, Sona-Azuero, southern
683
Panama and Chagres-Bayano, eastern Panama. Abbreviations in: (a) CR, Costa Rica,; GAA,
684
Greater Antilles Arc; LAA, Lesser Antilles Arc; PAN, Panama; PR, Puerto Rico. (b) CB,
685
Chagres-Bayano; CI, Coiba Island; Cord de Pan, Cordillera of Panama; GF, Golfito; PC, Panama
686
City; PCFZ, Panama Canal Fracture Zone.
687 688
Figure 2 Geologic map of the Azuero peninsula showing the location of our samples. Map is
689
modified from Dirección General de Recursos Minerales (1991), Buchs et al., (2010) and Corral
690
et al. (2011). See Figure 1 for broader location of the Azuero peninsula. Abbreviations: CB,
691
Chagres-Bayano; G, Group: GI, Group I; GII, Group II; GIII, Group III; Nic, Nicaragua; SA,
692
Sona-AzueroSee text for definitions of these groups.
693 694
Figure 3 MgO vs. SiO2 for all lavas of this study superimposed with the fields of MORB and
695
lavas of depleted arc systems (see text). Abbreivation: enr, enriched.
696 697
Figure 4 MgO vs. (a) TiO2, (b) Al2O3, (c) FeOt, (d) K2O, (e) Ce, (f) Yb, (g) Y and (h) Zr for all
698
lavas of this study superimposed with the fields of lavas of MORB and depleted arc systems
699
(basalts only)Abbreviation: enr, enriched.
700 701
Figure 5 Chondrite-normalized REE-normalized plots of Sona-Azuero lavas with 52 or less wt.
702
SiO2 from this study vs. Sona-Azuero, Golfito and Chagres-Bayano lavas with less than 52 wt. %
31
703
SiO2 from other studies. The similarity in chondrite-normalized LREE fractionations, i.e.,
704
[La/Yb]CN of our samples with those from other studies (Lissinna 2005; Buchs et al. 2010;
705
Wegner et al. 2011) in (a-c) form the basis for classification of our samples into three groups (see
706
Table 3 for further details): (i) Group I, LREE-depleted, (ii) Group II, LREE-enriched
707
(transitional to alkalic) and (iii) Group III, flat to slightly LREE-enriched. Chondrite
708
concentrations are from Nakamura (1974). Abbreviations: OP, oceanic plateau; PAR, plume- and
709
arc-related. The grey region in (b) to (f) represent the data presented in (a) and the dashed
710
regions in (d) to (f) represent the data presented in (c).
711 712
Figure 6 N-MORB-normalized plots of Sona-Azuero lavas with 52 or less wt. SiO2 from this
713
study vs. Sona-Azuero, Golfito and Chagres-Bayano lavas with less than 52 wt. % SiO2 from
714
other studies. N-MORB abundances are from Sun and McDonough (1989). Abbreviations: (a, e)
715
PAR, plume- and arc-related; (g, i) OP, oceanic plateau. The grey region in (b) to (f) represent
716
the data presented in (a) and the dashed regions in (d) to (f) represent the data presented in (c).
717 718
Figure 7 Zr vs. Zr/Y (Pearce and Norry 1979) relations of (a) subduction initiation rule ophiolite
719
(SIRO) basalts, (b) Sona-Azuero, Golfito and Chagres-Bayano basalts interpreted as oceanic
720
plateau and our Group I and II Azuero basalts and (c) Sona-Azuero, Golfito and Chagres-Bayano
721
basalts interpreted as arc and our Group III Azuero basalts. All Central American samples are
722
basalts only (i.e., no intrusives). Abbreviations in: (a) bas-and, basaltic andesite; DM, depleted
723
mantle; EM, enriched mantle; WPB, within-plate basalts. (b) Ha, Hauff et al. (2000a). (c) B,
724
Buchs et al. (2010).
725
32
726
Figure 8 (a) General petrogenetic paths followed by MORB and VAB and fields of MORB,
727
VAB and boninite (BON) in Y vs. Cr space and Y vs. Cr of (b) subduction initation rule
728
ophiolite basalts, (c) basalts of the Sona-Azuero, Golfito and Chagres-Bayano complexes
729
interpreted as oceanic plateau and our Groups I and II Azuero basalts and (d) arc basalts of the
730
Sona-Azuero, Chagres-Bayano and Golfito complexes and our Group III Azuero basalts (only,
731
no intrusives). All Central American samples are basalts only. The solid and dashed curves
732
represent incremental batch melting trends and source 1 (S1) and source 2 (S2) represent source
733
compositions from Murton (1989). S1 represents a calculated plagioclase lherzolite MORB
734
source comprised of 0.60 oli + 0.20 opx + 0.10 cpx and S2 represents the residue (residual
735
source) subsequent to 20% melt extraction from S1. The fields of MORB, VAB and within-plate
736
basalt (WPB) are from Pearce (1982) and the boninite (BON) field is from Dilek et al. (2007).
737
Abbreviations in: (b) bas-and, basaltic andesite; LU, lower unit; SIRO, subduction initiation rule
738
ophiolite; UU, upper unit. (c) enr, enriched; Ha, Hauff et al. (2000).
739 740
Figure 9 Ti/V (Shervais 1982) of (a) subduction initiation rule ophiolite (SIRO) basalts, (b)
741
Sona-Azuero, Golfito and Chagres-Bayano basalts interpreted as oceanic plateau and our Groups
742
I and II Azuero basalts and (c) Sona-Azuero, Golfito and Chagres-Bayano basalts interpreted as
743
arc and our Group III basalts. All Central American samples are basalts only (i.e., no intrusives).
744
Note that two of our enriched Sona-Azuero basalts fall off the plot to the right with 4.7-4.8 wt. %
745
TiO2 (~28000-29000 ppm Ti) and ~460 ppm V (Ti/V of ~53) in Fig. 9b, e within the within-
746
plate basalt field. The Izu Bonin Mariana FAB and Expedition 352 FAB are from Reagan et al.
747
(2010) and Shervais et al. (2019), respectively and the Expedition 351 FA-like basalts are from
33
748
Arculus et al. (2015). Abbreviations in: (a) BON, boninite; LU, lower unit, UU, upper unit. (b)
749
alk, alkaline; Ha, Hauff et al. (2000a); LU, UU, as in (a). (c) B, Buchs et al. (2010).
750 751
Figure 10a Chemotemporal relations between the Golfito, Sona-Azuero and Chagres-Bayano as
752
shown on the ‘synthetic tectonostratigraphic chart’ of Buchs et al. (2010). Unit acronyms (from
753
left to right): AP, Azuero (oceanic) Plateau (89-85 Ma); GA, Golfito Arc (75-66 Ma); SA, Sona-
754
Azuero Arc (75-39 Ma); CBA, Chagres-Bayano Arc (70-39 Ma). (b) Stratigraphy of an Azueru
755
section (the Rio Quema Formation) reproduced from Corral et al. (2011) which provides
756
relations between the Azuero oceanic ‘plateau’, proto-arc and arc. Other acronyms: PISI, plume-
757
induced subduction initiation. References for ages are provided in Table 1.
758 759
Figure 11 Two alternative tectonic models to reconcile subduction initiation at the Central
760
American Volcanic Arc system. (a) The earliest stages of subduction initiation were catalyzed by
761
plume emplacement (modified from Whattam and Stern 2015a). This event began to first form
762
the Sona-Azuero complex which was then followed by incipient formation of the Chagres-
763
Bayano complex, subsequent to slab rollback and hinge retreat. (b) Collision-induced subduction
764
initiation (see Buchs et al. 2010 for details) and subsequent arc construction upon a ~10 m.y.
765
oceanic plateau (based on the ideas of Buchs et al. 2010 and Wegner et al. 2011).
766 767
Table captions
768
Table 1 Age constraints for the Golfito, Sona-Azuero and Chagres-Bayano complexes.
769
Abbreviations: A, arc; CR, Costa Rica; PA, proto-arc; PAN, Panama; OP, oceanic plateau.
770
Superscripts: a, Azuero Arc includes the proto-arc of Buchs et al. (2010); b, is to indicate that
34
771
this refers to a restricted segment of the Azuero Arc (the Río Quema Formation, Corral et al.,
772
2011); c, Golfito also consists of basaltic trachyandesite and andesite according to Buchs et al.
773
(2010) and references therein; a TAS plot (not shown) does suggest this, however a plot of Nb/Y
774
vs. Zr/Ti (not shown) suggests that designation of alkaline affinities is the result of alteration
775
(addition of alkalis); d, similarly, the Nb/Y vs. Zr/Ti plot suggests no alkaline affinities; e,
776
Wegner et al. (2011) suggested alteration could account for the high K2O contents of some rock
777
types which is confirmed by a Nb/Y vs. Zr/Ti plot; f and g, interpreted by Hauff et al. (2000) and
778
Buchs et al. (2010) as oceanic plateau and proto-arc/arc, respectively.
779 780
Table 2 Whole rock ICP-AES analyzes of Groups I, II, and III Azuero lavas from this study.
781
Abbreviation: bas-and, basaltic-andesite.
782 783
Table 3 Mean and range of La/Yb normalized to chondrite ([La/Yb]CN) of our Group I, II and III
784
Sona-Azuero lavas vs. Sona-Azuero lavas of other studies (Buchs et al. 2010; Wegner et al.
785
2011). Chondrite abundances normalized to are from Sun and McDonough (1989).
786 787
References
788
Alabaster T., Pearce, J.A. and Malpas, J., 1982. The volcanic stratigraphy and petrogenesis of the
789
Oman ophiolite complex. Contributions to Mineralogy and Petrology 81, 168-183.
790
Allègre, C.J., Hart, S.R. and Minster, J.F., 1983. Chemical structure and evolution of the mantle
791
and continents determined by inversion of Nd and Sr idotopic data, II. Numerical experiments
792
and discussion. Earth and Planetary Science Letters 66, 191-213.
35
793
Allibon, J., Monjoie, P., Lapierre, H., Jaillard, E., Bussy, F., Bosch, D. and Senebier, F., 2008.
794
The contribution of the young Cretaceous Caribbean Oceanic Plateau to the genesis of late
795
Cretaceous arc magmatism in the Cordillera Occidental of Ecuador. Journal of South
796
American Earth Sciences 26, 355-368.
797 798 799 800
Arculus, R.J., 2003, Use and abuse of the terms calcalkaline and calcalkalic. Journal of Petrology 44, 929-935. Arculus, R.J. and the 29 others of IODP Expedition 351, 2015. A record of subduction initiation in the Izu-Bonin-Mariana arc. Nature Geoscience, http://dx.doi.org/10.1038/ngeo2515.
801
Barat, F., Mercier de Lépinay, B., Sosson, M., Müller, C., Baumgartner, P., and Baumgartner-
802
Mora, C., 2014. Transition from the Farallon Plate subduction to collision between South and
803
Central America: Geological evolution of the Panama isthmus. Tectonophysics 622, 145-167.
804
Beccaluva, L., Coltori, M., Giunta, G., and Siena, F., 2004. Tethyan vs. Cordilleran ophiolites: A
805
reappraisal of distinctive tectono-magmatic features of supra-subduction complexes in relation
806
to the subduction mode. Tectonophysics 39, 163-174.
807 808
Brounce, M., Kelley, K.A., Cotrell, E., and Reagan, M.K., 2015. Temporal evolution of mantle wedge oxygen fugacity during subduction initiation. Geology 43, 775-778.
809
Buchs, D.M., Arculus, R.J., Baumgartner, P.O., Baumgartner-Mora, C., and Ulianov, A., 2010.
810
Late Cretaceous arc development on the SW margin of the Caribbean Plate: Insights from
811
the Golfito, Costa Rica, and Azuero, Panama, complex. Geochemistry, Geophysics,
812
Geosystems, 11. doi:10.1029/2009GC002901.
813
Buchs, D.M., Baumgartner, P.O., Baumgartner‐Mora, C., Bandini, A.N., Jackett, S.‐J., Diserens,
814
M.‐O., and Stucki, J., 2009. Late Cretaceous to Miocene seamount accretion and mélange
815
formation in the Osa and Burica peninsulas (southern Costa Rica): Episodic growth of a 36
816
convergent margin, in The Origin and Evolution of the Caribbean Plate, edited by K. H.
817
James, M. A. Lorente, and J. L. Pindell, Geological Society Special Publication, v. 328, pp.
818
411–456.
819 820 821 822 823 824
Burov, E., and Cloetingh, S., 2010, Plume-like upper mantle instabilities drive subduction initiation: Geophysical Research Letters, v. 37, L03309, doi:10.1029/2009GL041535. Chauvel, C., and Blichert-Toft, J., 2001. A Hafnium isotope and trace element perspective on melting of the depleted mantle. Earth and Planetary Science Letters 190, 131-151. Cooper, P.A. and Taylor, B., 1985. Polarity reversal in the Solomon Islands arc, Nature 314, 428–430.
825
Corral, I., Griera, A., Gómez-Gras, D., Corbella, M., Canals, À., Pineda-Falconett, M., and
826
Cardellach, E., 2011. Geology of the Cerra Quema Au-Cu deposit (Azuero Peninsula,
827
Panama): Geologica Acta, 9(3-4), 481-498, doi : 1 0 . 1 3 4 4 / 1 0 5 . 0 0 0 0 0 1 7 4 2.
828
DeBari, S.M., Taylor, B., Spencer, K. and Fujioka, K., 1999. A trapped Philippine Sea plate
829
origin for MORB from the inner slope of the Izu-Bonin trench. Earth and Planetary Science
830
Letters 174, 183–197.
831 832
Dilek, Y., Furnes, H. and Shallo, M., 2007. Suprasubduction zone ophiolite formation along the periphery of Mesozoic Gondwana. Gondwana Research 11, 453-475.
833
Dilek Y., Furnes, H. and Shallo M., 2008. Geochemistry of the Jurassic Mirdita Ophiolite
834
(Albania) and the MORB to SSZ evolution of a marginal basin oceanic crust. Lithos 100,
835
174-209.
37
836
Elliot, T., Plank, T., Zindler, A., White, W., and Bourdon, B., 1997. Element transport from slab
837
to volcanic front at the Mariana arc. Journal of Geophysical Research 102, 14991–15019.
838
doi:10.1029/97JB00788.
839
Gazel, E., Carr, M.J., Hoernle, K., Feigenson, M.D., Hauff, F., Szymanski, D., and Van Den
840
Bogaard, P., 2009. The Galapagos-OIB signature in southern Central America: Mantle re-
841
fertilization
842
Q02S11,doi:10.1029/2008GC002246.
by
arc-hotspot
interaction.
Geochemistry,
Geophysics,
Geosystems,
843
Gazel, E., Hayes, J.L., Hoernle, K., Kelemen, P., Everson, E., Holbrook, W.S., Hauff, F., Van
844
Den Bogaard, P., Vance, E.A., Chu, S., Calvert, A.J., Carr, M.J., and Yogodzinski, G.M.,
845
2015. Continental Crust generated in oceanic arcs. Nature Geosciences 8, 321-327,
846
doi:10.1038/ngeo2392.
847
Geldmacher, J., Hanan, B. B., Blichert-Toft, J., Haarp, K., Hoernle, K., and Hauff, F., 2001.
848
Hafnium isotope variations in volcanic rocks from the Caribbean Large Igneous Province
849
and
850
doi:10.1029/2002GC000477.
Galápagos
hotspot
tracks.
Geochemistry,
Geophysics,
Geosystems
4,
851
Geldmacher, J., Hoernle, K., Van Den Bogaard, P., Hauff, F., and Klügel, A., 2008. Age and
852
geochemistry of the Central American forearc basement (DSDP Leg 67 and 84): Insights
853
into Mesozoic arc volcanism and seamount accretion on the fringe of the Caribbean LIP.
854
Journal of Petrology 49, 1781-1815, doi:10.1093/petrology/egn046.
855 856
Geldmacher, J., Heydolph, K., Murphy, D.T., Romanova, I., Mahoney, J. J. and K. Hoernle, 2012. American Geophysical Union, Fall Meeting 2012, abstract #DI51A-2339.
38
857
Gerya, T., Stern, R.J., Baes, M.. Sobolev, S.V. and Whattam. S.A., 2015. Plate tectonics on the
858
Earth
859
doi:10.1038/nature15752.
860 861 862 863
triggered
by
plume-induced
subduction
initiation:
Nature
527,
221-225,
Giudice, D. and Recchi, G., 1969. Geología del area del projecto minero de Azuero. Ejectora, informe tecnico. 48 pp., Gobierno de la Repub. de Panamá, Panama City. Goldstein, S. L., O’Nions, R.K., and Hamilton, P.J., 1988. Decoupled evolution of Nd and Sr isotopes in the continental crust. Nature 336, 733-738.
864
Hall, C.E., Gurnis, Sdrolias, M., Lavier, L.L. and Muellar R.D., 2003. Catastrophic initiation of
865
subduction following forced convergence across fracture zones, Earth and Planetary Science
866
Letters 212 15–30.
867
Hauff, F., Hoernle, K., Schmincke, H. –U., and Werner, R., 1997. A Mid-Cretaceous origin for
868
the Galapagos hotspot: Volcanological, petrological and geochemical evidence from Costa
869
Rican oceanic crustal fragments. Geol. Rundsch. 86, 141-155.
870
Hauff, F., Hoernle, K., and Schmidt A., 2003. Sr–Nd–Pb composition of Mesozoic Pacific
871
oceanic crust (site 1149 and 801, ODP Leg 185): Implications for alteration of ocean crust
872
and the input into the Izu–Bonin–Mariana subduction system. Geochemistry, Geophysics,
873
and Geosystems 4 (8).
874
Hauff, F., Hoernle, K.A., van den Bogaard, P., Alvarado, G.E. and Garbe‐Schonberg, D., 2000.
875
Age and geochemistry of basaltic complexes in western Costa Rica: Contributions to the
876
geotectonic evolution of Central America. Geochemistry, Geophysics, Geosystems 1, Paper
877
No. 1999GC000020.
39
878
Hawkesworth, C.J., Turner, S.P., McDermott, F., Peate, D.W. and van Calsteren, P., 1997. U-Th
879
isotopes in arc magmas: Implications for element transfer from the subducted crust. Science,
880
v. 276, p. 551-555.
881 882
Hoernle, K., Hauff, F., and van den Bogaard P., 2004. 70 m.y. history (139-69 Ma) for the Caribbean large igneous province. Geology 32, 697-700.
883
Hoernle, K., van den Bogaard, P., Werner, R., Lissinna, B., Hauff, F., Alvarado, G., and Garbe‐
884
Schonberg, D., 2002. Missing history (16–71 Ma) of the Galapagos hotspot: Implications for
885
the tectonic and biological evolution of the Americas. Geology 30, 795–798,
886
doi:10.1130/0091-7613(2002).
887
Ingle, S., Mahoney, J.J., Sato, H., Coffin, M.F., Kimura, J. –I., N. Hirano, N., and Nakanishi, M.,
888
2007. Depleted mantle wedge and sediment fingerprint in unusual basalts from the Manihiki
889
Plateau, central Pacific Ocean. Geology 35(7), 595-598, doi: 10.1130/G23741A.1.
890 891
Irvine, T.N., and Baragar, W.R.A., 1971. A guide to the chemical classification of the common volcanic rocks: Canadian Journal of Earth Sciences 8, 523-548, doi:10.1139/e71-055.
892
Ishizuka, O., Kimura, J.-I., Li, Yi B., Stern, R.J., Reagan, M.K., Taylor, R.N., Ohara, Y.,
893
Bloomer, S.H., Ishii, T., Hargrove III, U.S. and Haraguchi, S., 2006. Early stages in the
894
evolution of the Izu-Bonin arc volcanism: New age, chemical and isotopic constraints. Earth
895
and Planetary Science Letters 250, 385-401.
896
Ishizuka, O., Tani, K., Reagan, M.K., Kanayama, K., Umino, S., Harigane, Y., Sakamoto, I.,
897
Miyajima, Y., Yuasa, M., and Dunkley, D.J., 2011. The timescales of subduction initiation
898
and subsequent evolution of an oceanic island arc. Earth and Planetary Science Letters 306,
899
229-240.
40
900
Kerr, A.C., Marriner, G.F., Tarney, J., Nivia, A., Saunders, A.D., Thirlwall, M. J., and Sinton,
901
C.W., 1997. Cretaceous basaltic terranes in western Colombia: Elemental, chronological and
902
Sr–Nd isotopic constraints on petrogenesis. Journal of Petrology 38, 677-702.
903
Kerr, A.C., Tarney, J. Kempton, P.D., Spadea, P., Nivia, A., Marriner, G.F., and Duncan, R.A.,
904
2002. Pervasive mantle plume head heterogeneity: Evidence from the late Cretaceous
905
Caribbean-Colombian oceanic plateau. Journal of Geophysical Research 107(B7), doi:
906
10.1029/2001JB000790.
907 908
Kerr, A.C., White, R.V., and Saunders, A.D., 2000. LIP reading: Recognizing oceanic plateaux in the geological record. Journal of Petrology 41(7), 1041-1056.
909
Kerr, A.C., White, R.V., Thompson, P.M., Tamez, J., and Saunders, A.O., 2003. In: Bartolini,
910
C., Bufler, R.T., Blickwede, J. (Eds.), The Circum-Gulf of Mexico and the CaribbeanL
911
Hydrocarbon, habitats, basin formation and plate tectonics: American Associate of
912
petroleum Geologists (MPG) Memoir 79, pp. 126-188.
913
Kesler, S. E., Sutter, J.F., Issigonis, M.J., Jones, L.M., and Walker, R.L., 1977. Evolution of
914
porphyry copper mineralization in an oceanic island Panama. Economic Geology 72, 1142-
915
1153, doi:10.2113/gsecongeo.72. 6.1142.
916
Kolarsky, R.A., Mann, P., Monechi, S., Meyerhoff, H.D., and Pessagno Jr., E.A., 1995.
917
Stratigraphic development of southwestern Panama as determined from integration of
918
marine seismic data and onshore geology, in Geologic and Tectonic Development of the
919
Caribbean Plate Boundary in Southern Central America, edited by P. Mann, Special Paper,
920
Geological Society of America, v. 295, pp. 159-200.
921 922
Lissinna, B., Hoernle, K., and van den Bogaard, P., 2002. Northern migration of arc volcanism in western Panama: Evidence for subduction erosion? Eos Transactions, AGU 83, 1463–1464. 41
923
Lissinna, B., 2005. A profile though the Central American Landbridge in western Panama: 115
924
Ma interplay between the Galápagos Hotspot and the Central American Subduction Zone.
925
Ph.D. thesis, Christian-Albrechts University, Kiel, Germany, 102 pp.
926
Lissinna, B., Hoernle, K., Hauff, F., van den Bogaard, P., and Sadofsky, S., 2006. The
927
Panamanian island arc and Galápagos hotspot: A case study for the long-term evolution of
928
arc/hotspot interaction. Geophysical Research Abstracts 8, 05106.
929
Mamberti, M., Lapierre, H., Bosch, D., Jaillard, E., Ethien, R., Hernandez, J., and Polvé, M.,
930
2003. Accreted fragments of the Caribbean-Colombian Plateau in Ecuador. Lithos 66, 173-
931
199.
932 933 934 935
Marques, F.O., Cabral, F.R., Gerya, T.V., Zhu, G. and May, D.A., 2014. Subduction initiates at straight passive margins. Geology 42, 331-334. https://doi.org/10.1130/G35246.1. Matusomoto, T. and Tomoda, Y., 1983. Numerical simulation of the initiation of subduction at the fracture. Journal of Physics of the Earth 31, 183-194.
936
Maury, R.C., Defant, M.J., Bellon, H,. de Boer, Z.J., Stewart, R.H., and Cotten, J., 1995. Early
937
Tertiary arc volcanics from eastern Panama, in Geologic and Tectonic Development of the
938
Caribbean Plate Boundary in Southern Central America, edited by P. Mann, Specical Paper
939
of the Geological Society of America, 295, 29–34.
940
McCulloch, M.T. and Black, L.P., 1984. Sm–Nd isotopic systematics of Endreby land granulites
941
and
942
Planetary Science Letters 71, 46-58.
943 944
evidence for
the redistribution of Sm and Nd during metamorphism. Earth and
McCulloch, M.T. and Chappell, B.W., 1982. Nd isotopic charcteristics of S- and I-type granites. Earth and Planetary Science Letters 58, 51-64.
42
945 946
McDonough, W.F. and Sun, S.-S., 1995. The composition of the Earth, Chemical Geology 120, 223-253.
947
Meffre, S., Falloon, T. J. Crawford, T. J., Hoernle, K., Hauff, F. Duncan, R. A., Bloomer, S. H.
948
and Wright, D. J., 2012. Basalts erupted along the Tongan fore arc during subduction
949
initiation: Evidence from geochronology of dredged rocks from the Tonga fore arc and
950
trench.
951
doi:10.1029/2012GC004335.Metti, A., and Recchi, G., 1976. Geología de la Peninsula de
952
Soná e Isla de Coiba, in Memoria Segundo Congreso Latinoamericano de Geología, v. 2, pp.
953
541–553, Sucre, Caracas.
954 955
Geochemistry,
Geophysics,
Geosystems,
13,
Q12003,
Miyashiro, A., 1974. Volcanic rock series in island arcs and active continental margins. American Journal of Science 274, 321-355.
956
Montes, C., Bayona, G.A., Cardona, A., Buchs, D.M., Silva, C.A., Morón, S., Hoyos, N.,
957
Ramírez, D.A., Jaramillo, C.A., and Valencia, V., 2012a. Arc-continent collision and
958
orocline formation: Closing of the Central American Seaway. Journal of Geophysical
959
Research 117, B04105. doi.org/10.1029/2011JB008959.
960
Montes, C., Cardona, A., McFadden, R., Morón, S.E., Silva, C.A., Restrepo-Morono, S.,
961
Ramírez, D.A. Hoyos, N., Wilson, J., Farris, D., Bayona, G.A., Jaramillo, C.A., Valencia,
962
V., Bryan, J., and Flores, J.A., 2012b. Evidence for middle Eocene and younger land
963
emergence in central Panama: Implications for isthmus closure. Geological Society of
964
America Bulletin, doi:10.1130/B30528.1.
965 966 967
Nakamura, N., 1974. Determination of REE, Ba, Fe, Mg, Na and K in carbonaceous and ordinary meteorites. Geochimica et Cosmochimica Acta 38, 757-775. Pandey, D.K., Pandey, A. and Whattam, S.A., 2019. Relict subduction initiation along a passive 43
968
margin
969
https://doi.org/10.1038/s41467-019-10227-8.
in
the
northwest
Indian
Ocean.
Nature
Communications
10:2248,
970
Pearce, J.A., 1983. Role of the subcontinental lithosphere in magma genesis at active continental
971
margins, in Continental basalts and mantle xenoliths, edited by C. J. Hawkesworth and M.
972
J. Norry, pp. 230-249, Nantwich, Shiva.
973
Pearce, J.A., 2003. Supra-subduction zone ophiolites: The search for modern analogues, in
974
Ophiolite concept and the evolution of geological thought, edited by Y. Dilek, and S.
975
Newcomb, S., Geological Society of America Special Paper 373, pp. 269-293.
976 977
Pearce, J.A., 2008. Geochemical fingerprinting of oceanic basalts with applications to ophiolite classification and the search for Archean oceanic crust. Lithos 100, p. 14-48.
978
Pearce J.A, Lippard S. J., and Roberts, S., 1984. Characteristics and tectonic significance of
979
supra-subduction zone ophiolites, in Geology of Marginal Basins, edited by P. Kokelaar and
980
Howels, Geological Society of London Special Publication 16, pp. 77-94.
981
Pearce, J.A., Stern, R.J., Bloomer, S.H., and Fryer, P., 2005. Geochemical mapping of the
982
Mariana arc-basin system: Implications for the nature and distribution of subduction
983
components. Geochemistry, Geophysics, Geosystems 67, doi:10.1029/2004GC000895.
984 985
Pearce, J.A. and Norry, N.J., 1979. Petrogenetic implications of Ti, Zr, Y, and Nb variations in volcanic rocks. Contributions to Mineralogy and Petrology 69, 33-49.
986
Peucat, J.J., Vidal, P., Bernard-Griffiths, J., and Condie, K.C., 1988. Sr, Nd and Pb isotopic
987
systematics in the Archean low- to high-grade transition zone of southern India: Syn
988
accretion vs. post-accretion granulites. Journal of Geology 97, 537-550.
44
989
Phinney, E.J., Mann, P., Coffin, M.F., and Shipley, T.H., 1999. Sequence stratigraphy, structure,
990
and tectonic history of the southwestern Ontong Java Plateau adjacent to the North Solomon
991
Trench and Solomon Islands arc, Journal of Geophysical Research 104 (B9) 20446–20449.
992
Pindell, J.L., and Kennan, L., 2009. Tectonic evolution of the Gulf of Mexico, Caribbean and
993
northern South America in the mantle reference frame: an update, in The origin and
994
evolution of the Caribbean Plate, edited by K. H. James, M. A. Lorente, and J. L. Pindell,
995
328, London, Geological Society, pp. 1-55.
996
Portnyagin, M., Bindeman, I., Hoernle, K., and Hauff, F., 2007. Geochemistry of primitive lavas
997
of the Central Kamchatka Depression: Magma generation at the edge of the Pacific Plate, in
998
Volcanism and Subduction: The Kamchatka region, edited by J. Eichelberger, E. Gordeev,
999
M. Kasahara, P. Izbekov, and J. Lees, J., pp. 199-239. doi: 10.1029/172GM16.
1000
Reagan, M.K., Ishizuka, O., Stern, R.J., Kelley, K.A., Ohara, Y., Blichert-Toft, J,. Bloomer, S.–
1001
H., Cash, J., Fryer, P., Hanan, B.B., Hickey-Vargas, R., Ishii, T., Kimura, J.I., Peate, D.W.,
1002
Rowe, M.C. and Woods, M., 2010. Fore-arc basalts and subduction initiation in the Izu-
1003
Bonin-Mariana
1004
doi:10.1029/2009GC002871.
system:
Geochemistry,
Geophysics,
Geosystems
11,
Q03X12.
1005
Reagan, M.K., Pearce, J.A., Petronotis, K., Almeev, R.R., Avery, A.J., Carvallo, C., Chapman,
1006
T., Christeson, G.L., Ferré, E.C., Godard, M., Heaton, D.E., Kirchenbaur, M., Kurz, W.,
1007
Kutterolf, S., Li, H., Li, Y., Mishibayashi, K., Morgan, S., Nelson, W.R., Prytulak, J.,
1008
Python, M., Robertson, A.H.F., Ryan, J.G., Sager, W.W., Sakuyama, T., Shervais, J.W.,
1009
Shimizu, K., and Whattam, S.A. (2017) Subduction initiation and ophiolite crust: New
1010
insights
1011
http://dx.doi.org/10.1080/00206814.2016.1276482.
from
IODP
drilling.
45
International
Geology
Review,
1012
Robertson, A., 2004. Development of concepts concerning the genesis and emplacement of
1013
Tethyan ophiolites in the Eastern Mediterranean and Oman regions. Earth Science Reviews
1014
66, 331-387.
1015
Saccani, E., and Photiades, A., 2004. Mid-ocean ridge and supra-subduction affinities in the
1016
Pindos Massif ophiolites (Greece): Implications for magma genesis in a proto-forearc
1017
setting. Lithos 73, 229-53.
1018 1019
Salters, V.J.M., and Stracke, A., 2004. Composition of the depleted mantle. Geochemistry, Geophysics, Geosystems 5, (5), Q05004. doi:10.1029/2003GC000597.
1020
Shafaii Moghadam, H., Corfu, F., and Stern, R.J., 2013, U-Pb zircon ages of Late Cretaceous
1021
Nain-Dehshir ophiolites, central Iran: Journal of the Geological Society, v. 170, p. 175–184,
1022
https://doi .org /10 .1144 /jgs2012 -066.
1023 1024 1025 1026
Shaw, D.M., 1970. Trace element fractionation during anataxis. Geochimica et Cosmochimica Acta 34, 237-243. Shervais, J.W., 1982. Ti-V plots and the petrogenesis of modern and ophiolitic lavas. Earth and Planetary Science Letters 59, 110-118.
1027
Shervais, J.W., ,Reagan, M., , Haugen, E., , Almeev, R.R., , Pearce, J.A., , Prytulak, J., Ryan,
1028
J.G., , Whattam, S.A.,Godard, M., , Chapman, T., , Li, H., , Kurz, W.,, Nelson, W.R.,,
1029
Heaton, D.,, Kirchenbaur, M.Shimizu, K.,, Sakuyama, T.,Li, Y. , and Vetter, S.K., 2019.
1030
Magmatic Response to Subduction Initiation: Part 1. Fore-arc basalts of the Izu-Bonin Arc
1031
from IODP Expedition 352. Geochemistry, Geophysics, Geosystems, 20. https://doi.
1032
org/10.1029/2018GC007731
46
1033
Sinton, C.W., Duncan, R.A., and Denyer, P., 1997. Nicoya Peninsula, Costa Rica: A single suite
1034
of Caribbean oceanic plateau magmas. Journal of Geophysical Research 102, 15,507-15,520,
1035
doi:10.1029/97JB00681.
1036 1037
Stern, R.J., 2004. Subduction initiation: spontaneous and induced. Earth and Planetary Science Letters 226, 275-292.
1038
Stern, R.J. and Bloomer, S.H., 1992. Subduction zone infancy: examples from the Eocene Izu–
1039
Bonin–Mariana and Jurassic California. Geological Society of America Bulletin 104, 1621-
1040
1636.
1041
Stern, R.J. and Dumitru, T.A., 2019. Eocene initiation of the Cascadia subduction zone: A
1042
second
1043
https://doi.org/10.1130/GES02050.1.
1044 1045
example
of
plume-induced
subduction
initiation?
Geosphere.
Stern, R.J. and Gerya, T., 2018. Subduction initiation in nature and models: A review. Tectonophysics 746, 173-198.
1046
Stern, R.J., Reagan, M.,Ishizuka, O., Ohara, Y., and Whattam, S., 2012. To understand
1047
subduction initiation, study forearc crust: To understand forearc crust, study pphiolites.
1048
Lithosphere, doi: 10.1130/L183.1
1049
Sun, S.S. and McDonough, W.F., 1989. Chemical and isotopic systematic of oceanic basalts:
1050
Implications for mantle composition and processes, in Magmatism in the ocean basins,
1051
edited by A. D. Saunders and M. J. Norry, Geological Society of London Special Publication
1052
42, pp. 313-345.
1053
Tatsumi, Y. and Eggins, S., 1995. Subduction zone magmatism, Blackwell, London. 211 pp.
1054
Tomascak, P.B., Ryan, J.G., and Defant, M.J., 2000. Lithium isotope evidence for light element
1055
decoupling in the Panama subarc mantle. Geology 28, 507-510.
47
1056
Torró, L., Proenza, J.A., Marchesi, C., Garcia-Casco, A., Lewis, J.F., 2017. Petrogenesis of
1057
meta-volcanic rocks from the Maimón Formation (Dominican Republic): Geochemical
1058
record
1059
http://dx.doi.org/10.1016/j.lithos.2017.01.031.
of
the
nascent
Greater
Antilles
paleo-arc.
Lithos
278-281,
255-273,
1060
Tournon, J., Triboulet, C., and Azéma, J., 1989. Amphibolites from Panama: Anticlockwise P T
1061
paths from a pre-upper Cretaceous metamorphic basement in Isthmian Central America.
1062
Journal of Metamorphic Geology 7, 539–546, doi:10.1111/j.1525-1314.1989.tb00616.x.
1063 1064
Ueda, K., Gerya, T., and Sobolev, S.V., 2008, Subduction initiation by thermal-chemical plumes:
1065
Numerical studies: Physics of the Earth and Planetary Interiors, v. 171, p. 296–312,
1066
doi:10.1016/j .pepi.2008.06.032.
1067
Van der Lee, S., Regenauer-Lieb, K., and Yuen, D.A., 2008, The role of water in connecting past
1068
and future episodes of subduction: Earth and Planetary Science Letters, v. 273, p. 15–27,
1069
doi:10.1016/j .epsl.2008.04.041.
1070
Villagómez, D., Spikings, R., Magna, T., Kammer, A., Winkler, W., and Beltrán, A., 2011.
1071
Geochronology, geochemistry and tectonic evolution of the Western and Central cordilleras
1072
of Colombia. Lithos 125, 875-896.
1073
Wegner W., Wörner, G., Harmon R.S., and Jicha, B.R., 2011. Magmatic history and evolution of
1074
the Central American land bridge in Panama since Cretaceous times. Geological Society of
1075
America Bulletin 123, 703-734.
48
1076
Whattam, S.A., 2019. Primitive magmas in the early Central American Volcanic Arc system
1077
generated by plume-induced subduction initiation. Frontiers in Earth Science 6, 114, doi:
1078
10.3389/feart.2018.00114.
1079
Whattam, S.A., Malpas, J., Ali, J.R., Smith, I.E.M., 2008. New SW Pacific tectonic model:
1080
cyclical intraoceanic magmatic arc construction and near coeval emplacement along the
1081
Australia–Pacific margin in the Cenozoic. Geochemistry, Geophysics, Geosystems 9.
1082
doi:10.1029/2007GC001710.
1083
Whattam, S.A., Malpas, J., Smith, I.E.M., Ali, J.R., 2006. Link between SSZ ophiolite
1084
formation, emplacement and arc inception, Northland New Zealand: U–Pb SHRIMP
1085
constraints; Cenozoic SW Pacific tectonic implications. Earth and Planetary Science Letters
1086
250, 606–632.
1087
Whattam, S.A., Montes, C., McFadden, R.R., Cardona, A., Ramirez, D., and Valencia, V., 2012.
1088
Age and origin of earliest adakitic-like magmatism in Panama: Implications for the tectonic
1089
evolution of the Panamanian magmatic arc system. Lithos 142-143, 226-244.
1090
Whattam, S.A. and Stern, R.J., 2011. The ‘subduction initiation rule’: A key for linking
1091
ophiolites, intra-oceanic forearcs, and subduction initiation. Contributions to Mineralogy and
1092
Petrology 62, 1031-1045. doi: 10.1007/s00410-011-0638-z.
1093
Whattam, S.A. and Stern, R.J., 2015a. Late Cretaceous plume-induced subduction initiation
1094
along the southern margin of the Caribbean and NW South America: The first documented
1095
example with implications for the onset of plate tectonics. Gondwana Research 27, 38-63.
1096
Whattam, S.A. and Stern, R.J., 2015b. Arc magmatic evolution and the construction of
1097
continental crust at the Central American Volcanic Arc system. International Geology
1098
Review, http://dx.doi.org/10.1080/00206814.2015.1103668.
49
1099
Wilson, J.T., 1966. Did the Atlantic close and re-open? Nature 211, 676-681.
1100
Winchester, J.A. and Floyd, P.A., 1977. Geochemical discrimination of different magma series
1101 1102 1103
and their differentiation products using immobile elements: Chemical Geology 20, 325-343. Workman, R.K. and Hart, S.R., 2005. Major and trace element composition of the depleted MORB mantle (DMM): Earth and Planetary Science Letters 231(1-2), 53-72.
1104
Wörner, G., Harmon, R.S., and Wegner, W., 2009. Geochemical evolution of igneous rocks and
1105
changing magma sources during the evolution and closure of the Central American
1106
landbridge, in Backbone of the Americas: Shallow subduction, plateau uplift and ridge and
1107
terrane collision, edited by S. W. Kay, V. A. Ramos and W. Dickinson, Geological Society
1108
of America Memoir 204, pp.183-196.
1109
Zhou, X., Li, Z-H., Gerya, T.V., Stern, R.J., Xu, Z. and Zhang, J., 2018. Subduction initiation
1110
dynamics along a transform fault control trench curvature and ophiolite ages. Geology 46,
1111
607-610. https://doi.org/10.1130/G40154.1.
1112 1113
Zindler, A., and Hart, S.R., 1986. Chemical geodynamics. Annual Review of Earth and Planetary Sciences 14, 493-571.
50
Golfito Complex (i.e.,Proto-Arc, southern Costa Rica)
submarine
Azuero Complex (S. Panama) Azuero Plateau
submarine
Azuero Proto-Arc & Arc
Chagres-Bayano Arc (E. Panama)
Table 1 Whattam et al., 2019
submarine
?
subalkaline basalt, basaltic trachyandesite, trachyandesite
massive to thin pillowed lava flows
late Campanian to middle Maastrichian (~72-69 Ma)
biostratigraphic fossil ID
?
subalkaline basalt, (minor) gabbro
sheet flows, pillow basalts, rare gabbroic intrusives
Coniacian-Early Santonian (~89-85)
biostratigraphic fossil (radiolarite) ID
oceanic plateau
~1.7
tholeiitic basalt to rhyolite and their plutonic equilvalents (including enriched OIB-like varieties)
lavas flows, dykes, lava domes, large intrusive complexes
Late Cretaceous to Early Maastrichian 82-40
biostratigraphic fossil (foraminifera) ID;
proto-arc/arc
subalkaline basalt, (minor) gabbro basalt to rhyolite and their plutonic equilvalents
massive and pillowed lava flow-dyke complex with associated sub-volcanic plutonic rocks with overlying and intercalated volcaniclastic breccias
70-39
radiometric (40Ar/ 39Ar, LA-ICP-MS U-Pb on zircons)
~3
40
oceanic plateau and Dengo [1962]; Schmidt-Effing [1979]; Obando [1976]; DiMarco [1994]; DiMarco et al. [1995]; proto-arc/arc Hauff et al. [2000a]; Buchs et al. (2010)
Giudice and Recchi [1969]; Metti and Recchi [1976]; Kolarsky et al. [1995]; Hoernle et al. [2002]; Lissinna et al. [2002, 2006]; Lissinna [2005]; Buchs et al. [2009, 2010]; Wörner et al. [2009]; Corral et al. [2011]; Wegner et al. [2011]; Montes et al. [2012a]
39
radiometric ( Ar/ Ar, LA-ICP-MS U-Pb on zircons) arc
Maury et al. [1995]; Wörner et al. [2005, 2009]; Wegner et al. [2011]; Montes et al. [2012a, b]
Sample No. Lithology a Group
300070 basalt Group I
300071 basalt Group I
300075 basalt Group I
300076 basalt Group I
300077 basalt Group I
300078 basalt Group I
300079 basalt Group I
300081 basalt Group I
300082 basalt Group I
300082 (Rep) basalt Group I
300084 basalt Group I
300085 basalt Group I
LOI total
0.18 7.38 10.33 3.60 0.16 0.10 2.20 99.71
48.92 1.20 13.42 11.74 0.18 8.38 9.91 3.29 0.18 0.10 2.40 99.72
47.29 1.22 13.92 12.02 0.21 7.26 12.76 2.09 0.07 0.10 2.80 99.74
47.38 1.23 13.28 11.58 0.23 8.05 12.40 2.00 0.07 0.10 3.40 99.72
49.70 1.24 13.67 10.80 0.18 7.17 9.72 2.98 1.74 0.10 2.40 99.70
47.00 1.22 13.70 11.92 0.22 8.13 12.20 2.23 0.10 0.10 2.90 99.72
46.94 1.26 14.06 12.12 0.23 7.01 12.51 2.53 0.09 0.10 2.90 99.75
46.87 1.09 14.03 11.16 0.26 8.32 11.96 2.30 0.07 0.09 3.60 99.75
46.61 1.22 14.02 11.46 0.31 7.84 11.71 2.16 0.07 0.09 4.20 99.69
46.4 1.23 13.98 11.53 0.30 7.99 11.71 2.17 0.08 0.09 4.20 99.77
45.11 0.88 13.39 10.29 0.18 8.59 9.83 2.31 0.57 0.07 8.50 99.72
48.00 0.95 14.10 10.50 0.18 8.40 9.82 3.56 0.28 0.07 3.80 99.66
Mg#
56
59
55
58
57
58
54
60
58
58
62
61
48 316 46.4 138.1 67 2.4 204.8 21.6 58.7 3.9 <0.1 50 3.6
48 326 76.2 143.3 54 0.7 162.7 22.7 84.1 3.6 <0.1 14 3.2
50 331 69.6 142.8 61 0.6 118.1 22.6 61.8 3.7 <0.1 15 3.3
48 308 76.0 144.4 60 18.5 209.4 20.6 57.0 3.3 <0.1 204 2.8
51 336 69.4 140.1 49 1.1 155.4 22.4 58.2 3.2 <0.1 17 3.1
50 324 85.5 145.8 57 1.0 108.1 22.0 56.7 3.0 <0.1 17 3.0
48 289 65.8 128.3 50 0.7 142.2 18.7 66.1 2.9 <0.1 60 2.6
50 310 66.4 138.0 57 0.7 156.7 19.9 52.4 3.3 <0.1 27 2.8
50 314 NA NA NA 0.6 158.8 20.8 63.1 3.1 <0.1 31 2.9
46 275 51.5 109.7 49 1.1 134.9 16.0 41.8 3.2 <0.1 18 2.3
47 325 64.3 88.0 36 3.0 193.3 19.0 43.1 2.4 <0.1 151 2.5
major oxides (wt. %) SiO2 49.23 TiO2 1.22 Al2O3 13.55 Fe2O3 11.76 MnO MgO CaO Na2O K2O P2O5
trace elements (ppm) Sc 47 V 328 Ni 68.1 Cu 140.4 Zn 56 Rb 2.4 Sr 233.9 Y 21.9 Zr 59.2 Nb 3.3 Cs <0.1 Ba 37 3.2 La
Sample No. Lithology a Group
300086 basalt Group I
300088 basalt Group I
300089 basalt Group I
300091 basalt Group I
300072 basalt Group II
300073 basalt Group II
300074 basalt Group II
300083 basalt Group II
300066 andesite Group III
300067 basalt Group III
300068 basalt Group III
LOI total
0.20 8.10 11.78 2.39 0.14 0.10 2.40 99.69
47.93 1.25 16.75 11.04 0.20 6.40 9.48 3.29 0.67 0.11 2.60 99.72
48.78 1.76 13.72 13.18 0.21 6.23 10.80 2.74 0.17 0.18 2.00 99.77
47.04 1.26 13.86 11.74 0.21 7.55 12.64 1.80 0.04 0.10 3.50 99.74
44.77 4.55 11.78 16.96 0.23 5.52 8.34 3.46 0.46 0.44 3.20 99.71
43.97 4.69 11.82 16.68 0.21 5.46 11.06 2.15 0.51 0.45 2.70 99.70
46.95 2.76 13.34 14.18 0.22 6.51 9.82 3.25 0.33 0.24 2.10 99.70
45.00 3.64 13.09 13.37 0.17 5.62 9.46 3.86 0.92 0.43 4.20 99.76
56.28 1.14 15.37 9.05 0.17 3.46 6.37 3.94 0.99 0.31 2.80 99.88
48.11 0.57 20.30 8.79 0.15 4.91 10.80 3.23 0.47 0.07 2.40 99.80
47.06 0.51 18.78 10.11 0.19 7.24 12.99 1.63 0.10 0.05 1.10 99.76
Mg#
57
54
48
56
39
39
48
46
43
53
59
44 308 46.4 182.7 65 7.3 296.3 23.8 68.4 3.3 0.3 140 3.3
53 351 17.3 100.1 64 1.2 146.0 37.4 105.0 2.2 <0.1 71 4.2
48 309 73.6 135.4 52 0.4 172.0 21.4 63.3 3.8 <0.1 12 3.3
38 462 34.8 317.7 139 5.0 72.7 51.2 272.2 20.3 <0.1 51 17.4
38 463 34.6 323.7 107 7.1 155.9 53.4 281.5 20.9 <0.1 59 17.7
39 426 51.1 140.9 88 4.5 218.3 32.0 141.1 12.0 <0.1 92 9.8
28 336 56.0 25.1 75 16.5 473.0 29.5 237.6 32.7 <0.1 140 23.4
31 193 2.3 42.4 57 11.6 229.5 31.7 101.9 3.3 <0.1 397 10.9
33 241 16.3 151.5 40 7.2 373.7 11.6 25.5 1.2 0.3 159 2.6
43 296 9.9 221.6 17 2.1 360.7 10.7 11.0 0.8 0.1 72 2.3
major oxides (wt. %) SiO2 47.21 TiO2 1.22 Al2O3 14.03 Fe2O3 12.12 MnO MgO CaO Na2O K2O P2O5
trace elements (ppm) Sc 47 V 343 Ni 48.0 Cu 138.3 Zn 56 Rb 1.7 Sr 143.2 Y 23.0 Zr 59.1 Nb 3.6 Cs 0.1 Ba 60 3.2 La
Sample No. 300069 Lithology bas-and a Group Group III major oxides (wt. %) SiO2 54.30 TiO2 0.19 Al2O3 11.39 Fe2O3 11.49
300087 basalt Group III
300090 STD SO-18 STD SO-18 STD SO-18 STD SO-18 basalt Rep 1 Rep 2 Rep 3 Group III
LOI total
0.19 7.40 7.84 2.20 0.29 0.07 4.30 99.66
47.40 1.74 14.29 12.70 0.22 5.85 8.85 3.49 1.14 0.17 3.80 99.65
50.34 0.44 18.34 10.06 0.20 5.28 9.80 2.25 0.32 0.10 2.60 99.73
Mg#
56
48
51
47
47
47
47
44 336 19.2 107.8 75 14.5 336.9 32.9 100.2 2.5 <0.1 1100 5.0
41 267 9.1 118.1 35 8.2 764.3 13.5 25.6 0.5 0.1 219 3.4
26 192 46 NA NA 26.4 377.8 29 261.1 18.9 6.5 476 11.2
26 191 45 NA NA 26.4 380.7 29.7 264 19.1 6.7 484 11.4
25 205 46 NA NA 27.8 399 31.3 278.9 19.4 6.7 500 12.1
25 207 43 NA NA 28 413.7 31.4 282.7 20.5 7 513 12.2
MnO MgO CaO Na2O K2O P2O5
trace elements (ppm) Sc 59 V 338 Ni 20.3 Cu 71.7 Zn 128 Rb 3.2 Sr 164.8 Y 8.8 Zr 15.0 Nb 1.8 Cs <0.1 Ba 668 3.2 La
58.29 0.7 13.98 7.59 0.4 3.34 6.31 3.71 2.15 0.83 1.9 99.76
58.22 0.7 13.98 7.59 0.4 3.35 6.35 3.7 2.16 0.84 1.9 99.76
58.13 0.69 14.11 7.58 0.39 3.35 6.36 3.7 2.15 0.82 1.9 99.74
58.11 0.69 14.13 7.58 0.39 3.33 6.38 3.71 2.16 0.81 1.9 99.74
Ce
8.4
8.7
8.7
8.7
7.7
7.7
8.0
6.8
7.6
7.4
5.6
7.2
Ce
9.1
8.7
12.0
8.2
45.7
46.8
24.5
55
24.0
5.6
4.6
Ce
4.9
13.8
6.9
24.7
25.2
26.9
27.5
Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U
1.37 7.5 2.4 0.90 3.18 0.63 3.78 0.80 2.34 0.36 2.28 0.35 1.9 0.2 0.6 0.3 0.1
1.4 7.3 2.35 0.91 3.25 0.63 3.80 0.82 2.37 0.36 2.36 0.36 2 0.2 0.2 <0.2 0.1
1.42 7.1 2.51 0.97 3.39 0.64 3.76 0.82 2.44 0.36 2.30 0.36 2.3 0.2 0.6 0.3 0.1
1.41 7.4 2.47 0.92 3.40 0.64 3.86 0.83 2.41 0.35 2.33 0.35 2.0 0.3 0.2 0.2 0.2
1.32 6.5 2.19 0.90 2.97 0.59 3.77 0.79 2.30 0.35 2.35 0.34 2.0 0.2 0.2 0.4 0.1
1.35 7.5 2.35 0.92 3.24 0.62 3.90 0.82 2.40 0.37 2.30 0.33 1.8 0.2 0.3 <0.2 0.1
1.34 7.0 2.23 0.89 3.17 0.60 3.92 0.77 2.35 0.36 2.33 0.34 1.8 0.2 0.3 0.2 0.1
1.14 5.9 1.99 0.79 2.75 0.54 3.24 0.69 2.04 0.31 1.97 0.30 2.0 0.2 0.2 <0.2 <0.1
1.22 6.4 2.17 0.89 3.03 0.57 3.49 0.74 2.06 0.33 2.15 0.33 1.7 0.3 <0.1 <0.2 0.1
1.24 6.6 2.11 0.89 2.96 0.58 3.54 0.73 2.15 0.31 2.25 0.33 1.5 0.1 NA 0.2 0.1
0.97 5.1 1.62 0.63 2.27 0.44 2.71 0.59 1.69 0.26 1.84 0.26 1.2 0.1 0.1 0.2 <0.1
1.04 6.4 1.93 0.72 2.44 0.49 3.22 0.67 2.00 0.29 1.79 0.27 1.3 0.2 <0.1 <0.2 <0.1
Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U
1.42 7.1 2.14 0.99 3.14 0.61 3.66 0.85 2.62 0.34 2.29 0.34 1.8 0.2 0.3 <0.2 <0.1
1.39 7.4 2.41 0.92 3.37 0.65 4.22 0.89 2.68 0.39 2.55 0.39 1.9 0.2 0.2 0.4 0.1
2.13 12.6 4.00 1.41 5.44 1.04 6.42 1.38 4.11 0.62 4.03 0.59 3.1 0.1 0.8 0.3 <0.1
1.36 7.1 2.28 0.88 3.15 0.62 3.75 0.78 2.29 0.34 2.22 0.32 1.8 0.3 0.3 0.2 <0.1
6.87 34 9.23 2.83 10.75 1.76 10.06 1.84 5.02 0.68 4.36 0.60 7.7 1.5 1.1 1.5 0.5
7.31 34 9.74 2.91 10.94 1.83 10.05 1.97 5.22 0.71 4.65 0.62 8.5 1.3 1.1 1.3 0.6
3.77 18.1 5.33 1.85 6.21 1.06 6.07 1.20 3.24 0.47 3.07 0.42 4.0 0.8 0.6 0.8 0.4
7.27 33 7.21 2.39 7.47 1.11 6.00 1.05 2.80 0.37 2.39 0.33 6.3 2.1 2.1 1.8 0.5
3.51 16.4 4.28 1.37 5.23 0.89 5.63 1.16 3.43 0.51 3.51 0.50 3.1 0.2 1.8 1.9 0.6
0.89 4.2 1.29 0.57 1.74 0.33 2.02 0.44 1.32 0.21 1.31 0.20 0.9 <0.1 1.1 0.5 0.2
0.73 3.4 1.28 0.58 1.69 0.32 1.91 0.42 1.19 0.20 1.29 0.20 0.4 <0.1 1.3 <0.2 <0.1
Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U
0.90 4.2 1.05 0.35 1.24 0.22 1.26 0.28 0.85 0.15 1.05 0.18 0.5 <0.1 2.3 0.3 0.2
2.32 12.2 3.65 1.42 4.87 0.90 5.70 1.18 3.48 0.52 3.40 0.50 2.5 0.1 0.4 0.5 0.1
1.16 5.9 1.62 0.57 2.03 0.34 2.13 0.48 1.44 0.23 1.54 0.24 0.9 <0.1 1.3 0.4 0.2
3.15 13.1 2.73 0.81 2.77 0.48 2.74 0.58 1.71 0.27 1.63 0.27 8.8 6.8 NA 9.9 15.3
3.16 12.7 2.71 0.84 2.8 0.47 2.73 0.58 1.7 0.26 1.69 0.26 9 6.9 NA 9.9 15.4
3.27 13.6 2.84 0.84 2.95 0.5 2.86 0.6 1.71 0.27 1.73 0.26 9.2 6.9 NA 10.4 15.8
3.31 13.4 2.84 0.84 2.94 0.49 2.88 0.61 1.79 0.27 1.74 0.27 9.6 7.2 NA 10.1 16.1
a
Group refers to classification on the basis of petrographic characteristics (see Supplementary Document) and trace element chemistry (see Section 5.1 and Supplementary Document). Major element detection limits are about 0.001-0.4%. Other notes: Samples Group numbers in italics are samples with total oxide concentrations <96 wt. % and which were thus excluded from our chemical plots (see Section 4.3.2.). NA = Not analyzed. Rep = Replicate (analysis). STD = Standard. Table 2 Whattam et al., 2019
Table 3 Comparison of chondrite normalized La/Yb of Azuero samples with Sona-Azuero samples in the literature
Our data
Others
Group
mean [La/Yb]CN
range
n
mean [La/Yb]CN
range
n
I II III
0.96 2.63 1.43
0.75-1.09 2.29-2.86 1.28-1.58
12 3 3
0.92 2.39 1.43
0.63-1.15 1.96-2.81 0.76-3.21
36 3 17
Notes: The Group I of others comprise the Azuero and Sona 'CLIP' basalts of Wegner et al. [2011] and the Group I Azuero Plateau basalts of Buchs et al. [2010]; the Group II of others comprise the Group II basalts of Buchs et al. (2010); the Group III of others comprise arc basalts of Azuero, Sona and Coiba Island of Lissinna [2005] and the basalts of the proto-arc and arc of Buchs et al. [2010]. Chondrite abundances are from Sun and McDonough [1989].
Table 3 Whattam et al., 2019
100o W
o
o
90 W
70o W
80 W
NORTH AMERICAN PLATE Yucatan
60o W
N
Hispaniola
Cuba
Virgin Islands
Jamaica
CR 400 km
PR
GAA system
A s y s t em
CARIBBEAN PLATE
LA
Nicaragua
20o N
ean LIP ribb Ca Curacao
PAN
Fig. 1b
10oN
Venezuela
COCOS PLATE Colombia
a
Galapagos Islands
SOUTH AMERICAN PLATE
NAZCA PLATE
Caribbean Large Igneous Province (CLIP) PCFZ
08o N
PC
Pa n Fig. 2
Sona
Pearl Islands
Sona-Azuero Arc
Sona
CI
CB Arc
Majé
o
de
an
Cord
N
Chagre s-B ay
PANAMA Golfito GF Arc
b
CARIBBEAN PLATE
COSTA RICA 10 N o
o
0N
Bahia Pinia
Azuero Azuero
o
82 W
o
80 W
250 km
Figure 1 Whattam et al., 2019
a
Northwest
Southeast Caribbean coast
Pacific coast
eastern Panama
southern Costa Rica
southern Panama
Golfito Complex
Azuero Chagres-Bayano Complex Complex
1
PALEOGENE
Ma 40 50
70 80 90
Achiote Fm.
PALEOCENE Maastrichtian
L. CRETACEOUS
60
EOCENE
1
GA
Campanian Santonian Coniacian Turonian Cenomanian
100
CBA
~3 km
SAA
2
?
?
?
AP
AP
AP
? no sediment horizon or unconformity?
timing of PISI
“old plateau”
b NC
Rio Quema Formation Stratigraphic Section
calcarenite basaltic andesite dyke
v v
siltstone
Tonosi Formation meters
hemipelagic limestone
Azuero Arc
v
1500 +
+
+
+
+
+ v
v +
+
+
+
+
+
+
+
v
+
+
+
+
+
+
+
+
+
+
+
+
+
conglomerate
v +
+
+
+
+
+
+
+
+
+
+
+
+
dacite lava dome
+
+ + + + +
v v
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
crystal rich sandstone andesite basalt
v
+ + + + + + + + + + + + + + + +
v
+
v
+
+
v
+
v
v
+
v
+
v
v
+
v
+
+
v
+
+
+
v
+
Upper Unit +
v v v
+
v
+
+
v
+
+
v v v v v v
+
+
v
+
v
+
v v v v v v
+
+ + + + + + + + + +
+ + + + + + + + + +
+ + + + + + + + + + + + + + + + + + + +
v +
v
1000
turbidite
+
v
+ + + + + + + + + + + + + + + +
v
v v v v
v v v v v v v v
v v v v
v v v v
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
vv v v v v v v v
v
+
radiolarite
v
+
v
v v v v v v v v
500
v
pillow basalt
+
Limestone Unit + +
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ v
v v
v
v v
v
v v
v
v v
v
v v
v
v
0
+
v v v v v v
+ + +
Lower Unit
+
Proto-Arc Unit Azuero Igneous Basement (Oceanic Plateau)
Figure 10 Whattam et al. 2019
2
Golfito Formation Ocu Formation overlap sediments (forearc deposits) arc related igneous rocks & sediments proto-arc igneous rocks hemipelagic limestone oceanic plateau
a South
North zone of plumeweakened lithosphere BUOYANT
100-90 Ma: construction of CLIP Pacific
DENSE
thin oceanic crust
old thick lithosphere upwelling plume-modified asthenosphere
DENSE
mantle plume
85-75 Ma ?
lower, older Sona-Azuero OP & proto-arc/arc suite
thick OP crust of Azuero Plateau
Farallon Plate
proto-forearc spreading
stages of subduction initation
? thick OP crust of CLIP
Farallon Plate plu
me
flo
?
mantle plume(s)
90 Ma: plume-induced subduction initation
early
BUOYANT
89-85 Ma: construction of Azuero Plateau Pacific
thin oceanic crust
old thick lithosphere
b North
South
w
70 Ma partial melting of plume-contaminated MORB-like source subsequent to plume-induced subduction initiation (PISI); little to moderate fluid (shallow) subduction additions
plume-modified asthenosphere
75-73 Ma: collision-induced subduction initiation partial melting of plume-contaminated source ~ 10 m.y. subsequent to Azuero Plateau construction and subquent proto-arc and arc formation
upper, younger arc suite lower, older proto-arc suite Azuero Plateau
upper, younger Chagres-Bayano arc suite forearc spreading late hinge retreat
partial melting of an increasingly subduction-modified and deplteted source, via early shallow fluid flux hydration of mantle wedge & late (deep) sediment melt flux assisted melting subsequent to extraction of lavas and intrusives of the Sona-Azuero Complex
Figure 11 Whattam et al. 2019
o
15’
81 00’
45’
30’
o
15’
80 00’
50 km 300090 (GIII) o
8 45’
300066 (GIII) 300067 (GIII) 300091 (GI)
300068 (GIII)
300089 (GI)
300069 (GIII) 300088 (GI) 300071 (GI) 300070 (GI) 300072 (GI) 300073 (GII) 300076 (GI) 300079 (GI) 300078(GI) 300075(GI) 300077(GI) 300074(GII) 300086(I) 300087(GIII) 300081(GI) 300085(GI) 300082(GI) 300083(GII) 300084(GI)
30’
Playa Venado
Azuero peninsula
15’
Overlap sediments & alluvial deposits
CR GF Nic
C
oc
R os
Panama
CB
SA
id
ge
Azuero Accretionary Complex (Paleocene to Eocene) Accreted seamounts & oceanic islands Azuero Arc Group (Maastrichtian to Middle Eocene) Volcanic sequences & sediments & intrusives
Azuero
Azuero Proto-Arc Group (Late Campanian) Azuero Plateau (Albian? to Coniacian-Early Santonian)
Figure 2 Whattam et al. 2019
Sona-Azuero (89-85 Ma) others
Sona-Azuero this study
Grp I Grp II (enriched)
Grp I Grp II (enr)
Sona-Azuero Sona-Azuero this study (73-39 Ma) others Grp III Grp III proto-arc & arc
MORB depleted arcs
Golfito (Ha) (75-66 Ma) Chagres-Bayano (70-39 Ma) Golfito proto-arc (75-66 Ma, B)
Chagres-Bayano arc (70-39 Ma)
20
MgO wt.%
15
10
5
0 40
50
60
70
80
SiO2 wt.%
Figure 3 Whattam et al. 2019
Sona-Azuero (89-85 Ma) others
Sona-Azuero this study
Grp I Grp II (enriched)
Grp I Grp II (enr)
Sona-Azuero Sona-Azuero this study (73-39 Ma) others Grp III Grp III proto-arc & arc
MORB depleted arcs
Golfito (Ha) (75-66 Ma) Chagres-Bayano (70-39 Ma) Golfito proto-arc (75-66 Ma, B)
Chagres-Bayano arc (70-39 Ma)
5
25
a
b 20
Al2O3 wt.%
TiO2 wt.%
4
3
2
15
10
5
1
3
20
c
d K2O wt.%
FeOt wt.%
15
10
2
1
5
50
10
e
f Yb ppm
Ce ppm
40
30
20
5
10
300
100
g
h Zr ppm
Y ppm
80
60
40
200
100 20
0
0 0
5
10
15
20
MgO wt.%
Figure 4 Whattam et al. 2019
0
5
10
MgO wt.%
15
20
200
a
100
10
2
Sona-Azuero Group I 89-85 Ma plateau/CLIP (others) early hybrid PAR (this study)
200
b
100
10 Sona-Azuero Group II
2
89-85 Ma enriched plateau (others) enriched
200
c
Rock/Chondrites
100
10
Sona-Azuero Group III 73-39 Ma proto-arc & arc (others) later arc-dominated PAR (this study)
200 Sona-Azuero
100
CLIP & our Gr I basalts
d
arc & our Gr III basalts
10
Golfito Complex (75-66 Ma) interpreted as proto-arc interpreted as CLIP OP
2 200
e
100
10
2
Chagres-Bayano Complex interpreted as CLIP OP
200 100
Chagres-Bayano Complex interpreted as arc
f
10
1 La Ce Pr Nd
Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
Fig.5 Whattam et al. 2018
400
a
100
Sona-Azuero Group I 89-85 Ma plateau/CLIP (others) early hybrid PAR (this study)
10
1
.1 400
b
100
10
1 Sona-Azuero Group II
.1
89-85 Ma enriched plateau (others) enriched
400
c
100
Sona-Azuero Group III 73-39 Ma proto-arc & arc (others) later arc-dominated PAR (this study)
Rock/N-MORB
10
1
.1 400
d
100
Golfito Complex (75-66 Ma)
10
interpreted as proto-arc interpreted as CLIP OP
1
.1 400
e
100 Chagres-Bayano Complex
interpreted as CLIP OP
10
1
.1 400
f
100 Chagres-Bayano Complex interpreted as arc
10
1
.1 Rb Th Nb La Pb Sr Nd Sm Ti Y Lu Cs Ba U K Ce Pr P Zr Eu Dy Yb
Fig. 6 Whattam et al. 2019
20
a 84% SIRO LU
100% SIRO UU
10
WPB
EM
DM
MORB
VAB 1
SIRO lower unit (LU) basalt SIRO upper unit (UU) bas-and
20 Sona-Azuero (89-85 Ma) others
10
Sona-Azuero this study Grp I Grp II (enr)
b
Golfito (Ha) (75-66 Ma) Chagres-Bayano (70-39 Ma)
Zr/Y
Grp I Grp II (enriched)
1
20 10
Sona-Azuero Sona-Azuero (73-39 Ma) others this study Grp III Grp III proto-arc & arc
Golfito proto-arc (75-66 Ma, B)
Chagres-Bayano arc (70-39 Ma)
c
all tholeiitic basalts (from b)
1 .5 10
100
500
Zr ppm
Figure 7 Whattam et al. 2019
7000
S2
S1
40%
1000
40%
to S3
20%
20%
10%
relative crystallization sequence of plag & cpx MORB: plag cpx VAB: cpx plag 20%
10%
40%
BON
10%
MORB fra
cti
fr a c
100
t io n a t io
Petrogeneses of MORB & VAB
on
ati
on
VAB
n
a
7000 S2
S1
40%
1000
SIRO LU basalt SIRO UU bas-and
40%
to S3
20%
10%
10%
20%
20%
40%
10%
Cr ppm
100
b
40%
7000
S2
S1
40%
1000
40%
to S3
20%
20%
10%
10%
20%
Golfito OP (Ha) (75-66 Ma) Chagres-Bayano OP
40% 10%
100 Sona-Azuero OP (89-85 Ma) others
Sona-Azuero this study
Grp I Grp II (enr)
7000
Grp I Grp II (enr)
S2
S1
40%
1000
40%
to S3
20%
20%
10%
30%
c
40%
Sona-Azuero proto-arc & arc (73-39 Ma) others Grp III
Sona-Azuero this study Grp III
20% 40% 10%
100 Golfito proto-arc (75-66 Ma) Chagres-Bayano arc (70-39 Ma)
40%
d
10 3
10
100
Y ppm
Figure 8 Whattam et al. 2019
Comparison with SIRO lavas
Comparison with FAB
BON
BON
ARC
600
V Ti/
=2
Izu-Bonin Mariana FAB Exp. 352 FAB Exp. 351 FA-like basalts
Ti/ V
SIRO LU basalt SIRO UU bas-andesite
Ti/ V
800
=1
=1
0
0
1000
ARC
0
MORB
400
= Ti/V
V Ti/
=2
0
MORB
50
= Ti/V
WPB
200
50
WPB
a
d
Golfito OP (Ha) (75-66 Ma)
Grp I Grp II (enr)
Chagres-Bayano OP
BON
SIRO UU
600
ARC
SIRO LU
V Ti/
=2
0
MORB
400
= Ti/V
50
WPB
200
e
b =1
0
1000 Sona-Azuero proto-arc & arc (73-39 Ma) others Grp III
800
Ti/ V
V ppm
Sona-Azuero this study
Grp I Grp II (enriched)
Ti/ V
Sona-Azuero OP (89-85 Ma) others
800
=1
0
1000
Sona-Azuero this study Grp III
600
ARC V Ti/
BON
Golfito proto-arc (75-66 Ma, B)
=
20
MORB
Chagres-Bayano arc (70-39 Ma)
400
= Ti/V
50
WPB
200
Group I (from b)
0
5
10
15
f
c
0 20
25 0
Ti ppm/1000
Figure 9 Whattam et al. 2019
5
10
15
20
25
Highlights
• • •
Older Sona-Azuero complex is MORB-like Younger Chagres-Bayano complex is arc-like Early Central American forearc adheres to SIR