Along-Strike Geochemical Varia�ons in the Late Triassic Nikolai Magma�c System, Wrangellia, Central Alaska
Alicja WYPYCH¹ ([email protected]), Evan Twelker¹, Lauren Lande¹,², Rainer Newberry²¹Alaska Division of Geological & Geophysical Surveys, ²University of Alaska Fairbanks
The Triassic Nikolai Basalt and related mafic to ultramafic intrusions are one of the world’s most complete and best exposed sec�ons of a large igneous province (Amphitheater Mountains, Alaska), and have been explored for magma�c Ni-Cu-Co-PGE mineraliza�on (Wellgreen deposit, Kluane Ranges, Yukon Territory, and Eureka zone, Eastern Alaska Range, Alaska). The full extent of the basalts and intrusions, as well as magma genera�on processes are yet to be fully established. Two hypotheses have been put forward to explain the geochemical varia�on in Nikolai Basalts: 1) a deep mantle plume (Greene et al., 2008), and 2) decompressional mel�ng of mantle beneath a back-arc/intra-arc spreading center (Taylor et al., 2008). To be�er understand the extent and magma�c architecture of the Nikolai system, inves�gate the mineral poten�al and explore the two hypotheses of basalt genera�on, the Alaska Division of Geological & Geophysical Surveys (DGGS) conducted mapping and geochemical inves�ga�ons of the province from 2013 through 2015.We present major and trace element data from whole rocks and olivine from Triassic basalts and intrusive rocks collected over a 250 km along-strike transect. Chemical composi�ons of whole rock samples were obtained using induc�vely coupled plasma atomic emission spectroscopy and induc�vely coupled plasma mass spectrometry. This data is used to answer ques�ons about varia�ons in magma genera�on, temperature of crystalliza�on, and degree of frac�onal crystalliza�on required to produce the Nikolai Basalts.
INTRODUCTION
130°0'0"W
130°0'0"W
140°0'0"W
140°0'0"W
150°0'0"W
150°0'0"W
160°0'0"W
160°0'0"W
65
°0'0
"N
65
°0'0
"N
60
°0'0
"N
60
°0'0
"N
55
°0'0
"N
55
°0'0
"N
50
°0'0
"N
50
°0'0
"N0 600 1,200 1,800 2,400300Kilometers
Alexander
Wrangellia
Peninsular
Juneau
Fairbanks
Anchorage
¹
FGFGFGF GFGFGF F
FFGFFGFFFFGFGFGFGF
FFF
GFGF
GFFFFFFF
F
F
FFGF
FFF FFF
GFGFGFGFGF
GFGFGFGF
GFFFGFFGFF
GFGFGFGFFFFGFFFGFF
GFFFGFFFGFGFGFGFGFGFGFGFGFGFGFGFGFFFFFF
FFFFFFFFFFFGFFFFFFFF
FF
GFGFGFFGFGF
F
FGFGFF
GFGFGFGFGF
GFGF
GFFGFGFGFGFGFFGFGFFF
GFGF
GFGFGFFGFFGFFGFFGFGF FFFFF
GFF
GF
FF
FFFF
F
GF
F
F
GF
GF
F
GF
F
GF
FGF
F
GFGF
FFGF
GFGFF
GFGF FGFGF GFGFGFF
FF GFGF
FGFFF
FFFF
GFFGFGF FFFF
GFGFGF FGF
F
GFGFGFGF
GF FFFGFFF
FGFGFGF
GFGFGFGFGFFFGFFF
GF GFGFF
GFGF
GFGF
GFGF
GF
GFGFGF
GF
FFGFF
GF
FFF
FF FFF
FF
FFFGFFFFFFFF FGF GFFFGFFFF
FFFFGFFF
F
FFFFFFF FFGF
FGFGFFFFFF
GFFFGFGFGFGF GF
GF
146°0'0"W
146°0'0"W
147°0'0"W
147°0'0"W
148°0'0"W
148°0'0"W
149°0'0"W
149°0'0"W
150°0'0"W
63
°0'0
"N
63
°0'0
"N6
2°0
'0"N
145°0'0"W
62
°0'0
"N
Wrangellia
Peninsular-Talkeetna
Geochemical classifica�on
G high Ti mafic/ultramafic volcanic
G high Ti mafic/ultramafic intrusions
F low Ti mafic/ultramafic volcanic
F low Ti mafic/ultramafic intrusions
¹
Terranes
WrangelliaPeninsular
0 25 50 75 10012.5Kilometers
Limestone and marble
Metasediments and metavolcanicsNikolai Greenstone, upper member
Nikolai Greenstone, lower
mafic, ultramafic intrusions, undifferen�ated
mafic, ultramafic intrusions, upper
mafic, ultramafic intrusions, lower
Triassic Units
146°0'0"W
146°0'0"W
147°0'0"W
147°0'0"W
148°0'0"W
148°0'0"W
149°0'0"W
149°0'0"W
150°0'0"W
63°0
'0"N
63°0'0"N62°0'0"N
145°0'0"W
62°0
'0"N
Wrangellia
Peninsular-Talkeetna
¹
Terranes
WrangelliaPeninsular
Geochemical classifica�onultramafic intrusions with 18 wt% < MgO > 12 wt% ultramafic intrusions with MgO > 18 wt%
picriteultramafic volcanic rocks with MgO > 18 wt%ultramafic volcanic rocksmafic volcanic rocksmafic intrusions
0 25 50 75 10012.5Kilometers
Limestone and marble
Metasedimentery and metavolcanic rocksNikolai Greenstone, upper member
Nikolai Greenstone, lower member
mafic/ultramafic intrusions, undifferen�ated
mafic/ultramafic intrusions, upper
mafic/ultramafic intrusions, lower
Triassic Units
Fig. 3. Geographic distribu�on of geochemically differen�ated rocks in the map area. The geochemical classifica�on of mafic and ultramafic rocks is based on silica, potassium, sodium and magnesium oxide composi�on (LeBas et al., 1986; 2000). The larger the circle, the more magnesium rich the rock. The high and very high Mg rocks tend to concentrate in the NE corner of the map area, and dissipate towards the SW.
Fig. 2 Rock samples classified based on �tanium and zirconium composi�on (Pearce and Cann, 1973). High TiO₂ volcanic rocks lie stra�graphically on top of low TiO₂ volcanicrocks, and generally have greater thickness (Twelker et al., 2015), however the geographic distribu�on of both is similar.
Fig. 1. A) Geographic distribu�on of Triassic mafic and ultramafic rocks (green dots represent data collected by Taylor et al. [2008], Greene et al. [2008, 2009]), the blue, pink and maroon dots represent samples collected by DGGS during 2013-2015 field seasons (Twelker et al., 2014; Wypych et al., 2014; Wypych et al., 2015). Pink, red and green shaded polygons represent Alexander, Wrangellia and Peninsular Terranes (USGS data), the orange rectangle outlines the area of focus of this study. The Nikolai Basalts in the region of interest erupted onto the Skolai Arc (Nokleberg et al., 1994).B) Schema�c cross sec�on through Paleozoic and Triassic metavolcanic and metasedimentary rocks in the study area. Skolai Arc in the region is build mostly with andesi�c metavolcaniclas�c rocks with some rhyoli�c layers, and is capped by limestone and marble. Overlying low TiO₂ Nikolai (Late Triassic) and High TiO₂ Nikolai are largely metabasalts and meta-basal�c andesites. The mafic and ultramafic intrusions (low and high TiO₂) are interpreted as feeders to the Nikolai basalts.
B)A)
C)
D)
Fig. 4. A) Winchester and Floyd (1977) geochemical classifica�on of Nikolai volcanic rocks. Two groups (open and closed symbols) represent low- and high-TiO₂ composi�on of the basalts (lower and upper group). Lower group basalts tend to plot within the basalt to subalkaline basalt field, whereas upper group basalts plot solely in the subalkaline basalt field. The upper and lower groups are also dis�nguished by thair V and Zr composi�on; the upper group plot within MORB fields both on the Shervais (1982) diagram (fig. B), and Pearce and Cann (1973; Ti vs. Zr plot, not shown), and the lower group plot mostly within Island Arc Tholeiites. C) On Pearce`s (2008) diagram, samples from the upper group plot �ghtly within enriched MORB field, whereas the lower group samples sca�er mostly above the MORB array sugges�ng addi�onal processes involved in the forma�on of those rocks. D) Zr-Ti-Y diagram (Pearce and Cann, 1973) shows samples from upper group displaying within-plate classifica�on, and the lower group sca�er mostly towards island-arc/ocean floor fields.
Green polygons outline upper and lower Nikolai basalts analyzed by Taylor et al. (2008) and orange polygons outline data collected by Greene et al. (2005; 2008; 2009). DGGS data geerally falls within their composi�onal polygons.
Zr Yx3
Ti/100
C
D A
B
Island-arc A B
Ocean-floor B
Calc-alkali B C
Within-plate D
0.1 1 10 1000.01
0.1
1
10
Th/Yb
Nb/Yb
NM
ORB
EMO
RB
OIB
0 5 10 15 20 250
100
200
300
400
500
600V (ppm)
Ti (ppm)/1000
Low-Ti IATBON
Island Arc tholeiite (IAT)
MORB
Alkaline
10 ARC<20>OFB
50
1000.01 0.1 1
0.001
0.01
Zr/TiO₂
Nb/Y
SubAlkaline Basalt
Andesite/Basalt
Andesite
Alk-B
as
0 1 2 3 4 50
10
20
30
40
50
Ce
Yb0 1 2 3 4 5
0
10
20
30
40
50
Ce
Yb
0.001 0.01 0.1 10.0001
0.001
0.01
0.1
1
Pt
Pd
chondrite Pt/Pd = 1.9
0 10 20 30 40 500
10
20
30
40
Ce
Y
EMORB
SKOLAI65
75
80
85
90% AFC
mod
el 1
DMORB 3045
70
75
80% AFCmodel 2
0.0
0.5
1.0
1.5
2.0
Th
40
55
65
75
85% AFC
mod
el 1
DMORB20
35
50
65
75
80% AFCmodel 2
EMORB
SKOLAI
0 20 40 60 80 100 120 1400
1
2
3
4
Sm
Zr
DMORB
65
75
80% AFC
model 2
1525
35
45
55
65
75
85% AFCmodel 1EMORB
SKOLAI
0
1
2
3
TiO₂
DMORB20
35
65
75
80
85% AFCmodel 2
55
65
70
75
80
85% AFCmodel 1EMORB
SKOLAI
Fig. 6 Assimila�on - Frac�onal Crystalliza�on (AFC) Models (DePaolo, 1981). Since normal EMORB mel�ng and frac�onal crystalliza�on cannot explain the low TiO₂ Nikolai group we conducted trace-element modeling. The models presented here are the best fit to explain our data. Both models were run with depleted MORB (DMORB; Workman and Hart, 2005) as parental magma and average composi�on of Skolai Arc (red circle; Wypych et al., 2014) as assimilant, with a frac�onal crystalliza�on to assimila�on ra�o of 3.28 and at 1 atm pressure. In model 1 we used 20% olivine, 65% of clinopyroxene and 15% feldspar as frac�onal crystalliza�on material and in model 2 we crystallized 65% olivine, 20% clinopyroxene and 15% feldspar - both the composi�ons are present within the most mafic intrusions in the region. Other experiments (simple frac�ona�on of MORB, EMORB, DMORB, and AFC of EMORB and Skolai Arc) have been conducted, however none of those was able to simulate the chemical composi�on of the Nikolai groups. The models are plo�ed on bivariate plots ( Th vs. Y, Ce vs. Y, TiO₂ vs. Zr and Sm vs. Zr) with average low TiO₂ and high TiO₂ groups and cumulates. 50 to 60 % of assimila�on and frac�onal crystalliza�on of the experiment with crystalliza�on of 65% olivine, 20% of clinopyroxene and 15% of feldspar best explain the low TiO₂ Nikolai composi�on. The high TiO₂ Nikolai group can be explained by both experiments, however model 1 is a slightly be�er fit.
GEOCHEMISTRY OF THE TRIASSIC MAFIC AND ULTRAMAFIC ROCKS
GEOCHEMICAL CLASSIFICATION OF MAFIC VOLCANIC AND PLUTONIC ROCKS
High TiO₂ volcanic rocks
High TiO₂ intrusive rocks
Low TiO₂ volcanic rocks
Low TiO₂ intrusive rocks
GEOCHEMISTRY OF TRIASSIC MAFIC AND ULTRAMAFIC ROCKS PRELIMINARY MAJOR AND TRACE ELEMENT MODELING
A)
B)
C)
D)
Fig. 5 Preliminary major element modeling using pMELTS program (Ghiorso et al., 2005). To inves�gate the origin of the Nikolai magmas we undertook pMELTS modeling; here we show experiments conducted at 500 bars pressure, with star�ng temperature of 1200º C and 50º C steps. Since the high TiO₂ group of Nikolai basalts and intrusions plot in the EMORB field (fig. 4 B, and C) we used EMORB composi�on as a star�ng point (composi�ons from Waters et al., 2010). We find that the high TiO₂ group can easily be explained by simple par�al mantle mel�ng followed by about 10 to 20% frac�onal crystalliza�on, however this mantle composi�on is not capable of producing the low TiO₂ group Nikolai Basalt. The diamonds tend to plot as more mafic than the volcanic and intrusive “normal groups”. They have been flagged as different because of their very high MgO content, and are interpreted by us as cumulates from the frac�onal crystalliza�on of the Nikolai magma. The major element frac�onal crystalliza�on pa�erns seem to support this interpreta�on (especially diagrams B and D). To further inves�gate the theory that the very high MgO samples are in fact cumulates, we performed a similar MELTS experiment using the most mafic basalt composi�on both for high TiO₂ and low TiO₂ group Nikolai as star�ng liquid composi�on (insert into diagram 5 B) and plo�ng olivine composi�on crystallizing from that liquid composi�on against olivine composi�on measured by electron microprobe with wavelength dispersive X-ray spectroscopy. Olivines from the cumulates always plot as more mafic than the high TiO₂ Nikolai group and fall along the sample frac�onal crystalliza�on trend, but cannot be explained by frac�onal crystalliza�on of the low TiO₂ group.
1207ºC
1125ºC
1251ºC
1207ºC
1157ºC
1207ºC
1157ºC
1207ºC
1107ºC
0.5 0.6 0.7 0.8 0.9 1.035
37
39
41
SiO₂
Forsterite
frac�
onal cry
stalliz
a�on
Average of upper group volcanic rocks
Average of upper group intrusive rocks
Average of lower group volcanic rocks
Average of lower group intrusive rocks
AFC model
Olivine from
cumulates
PLATINIUM GROUP PROVENANCING
Fig. 7 A) Bivariate sca�er plot showing Ce vs. Yb for upper and lower groups. Rocks of the upper magma�c suite plot along a posi�vely sloped trend with higher Ce:Yb ra�os, reflec�ng their characteris�c enriched REE pa�ern. The rocks of the lower suite generally have lower Ce:Yb ra�o, and sca�er similarly to fig.4. Olivine cumulates have low total REE values because REE are incompa�ble in olivine; nevertheless, many of the cumulate rocks retain the Ce:Yb ra�o of the high TiO₂ magma�c suite. B) The same plot showing samples with elevated values of Pt and Pd. These data indicate that PGE mineraliza�on occurs in cumulate rocks associated with upper Nikolai group magmas.
Upper group volcanic rocks
Upper group intrusive rocks
Lower group volcanic rocks
Lower group intrusive rocks
Picrites
Cumulates
Samples with elevated Pt and Pd
Fig. 8. Bivariate log-log sca�er plot showing Pt vs. Pd values. The data show two dis�nct popula�ons with differing Pt:Pd ra�os. High TiO₂ Nikolai extrusive and intrusive rocks have �ghtly clustered Pt and Pd values; however, picrites and olivine cumulates have a higher Pt:Pd ra�o that is indis�nguishable from that of the lower group. Differing characteris�c Pt:Pd ra�os may reflect the condi�ons of mantle mel�ng or magma ascent.
DISCUSION AND CONCLUSIONS
1) Trace-element composi�on points toward enriched MORB as a source for high TiO₂ Nikolai group2) pMELTS modeling confirms the possibility of the high TiO₂ group being derived from a mantle source, however it cannot explain the low TiO₂ group sufficiently3) Trace element modeling indicates that deep mantle par�al melts mixing with Skolai Arc are required to produce the low TiO₂ Nikolai composi�on and can also explain the high TiO₂ Nikolai group 4) Elevated pla�num group elements are associated with high TiO� ultramafic, high MgO olivine cumulates of the high TiO₂ Nikolai group. 5) The Pt/Pd composi�on of the high TiO₂ Nikolai group could suggest scavenging of the PGE from the low TiO₂ Nikolai by the ascending high TiO₂ Nikolai melts.
Two models have been proposed to explain the two groups within the Nikolai system. Greene et al. (2008) proposed a deep plume as the source of the volcanic rocks whereas Taylor et al. (2008) proposed decompressional mel�ng of mantle due to intra- or back-arc ri�ing. Our geochemical modeling indicates the necessity of a depleted upper mantle component mixed with Skolai Arc to produce the low TiO₂ Nikolai member. The mixing with inherently heterogeneous arc could explain the sca�ered geochemical behavior of the low TiO₂ Nikolai samples. The �ghtness and homogeneity of the high TiO₂ Nikolai suggest that the homogeneous upper mantle was solely the source of those melts. The geochemical data could support both of the proposed models, however the geographic extent of the Nikolai groups following the arc suggests the back-arc/intra-arc spreading and decompressional mel�ng is a more plausible scenario.
Origin of the low TiO₂ and high TiO₂ Nikolai groups:
ACKNOWLEDGMENTSThis project is funded by the Legislature as part of the Governor’s Strategic and Cri�cal Minerals Assessment Capital Improvement Project which supplements the Airborne Geophysical/Geological Mineral Inventory as well as USGS Na�onal Coopera�ve Geologic Mapping Program STATEMAP component (award number G14AC00167 for 2014)
REFERENCES
8
9
10
11
12
13
14
FeO*Average of upper group volcanic rocks
Average of upper group intrusive rocks
Average of lower group volcanic rocks
Average of lower group intrusive rocks
Average of picrites
Average of cumulates
pMELTS model
0 1 2 30
10
20
MgO
TiO₂
7
8
9
10
11
12
13
14
15
CaO
44 46 48 50 52 540
1
2
3
TiO₂
SiO₂
liquid composi�on
liquid composi�on
liqui
d co
mpo
si�o
n
liquid composi�on
vv
vv v
v
v
v
vvvv
v
vv v
vv
v
v v
vvv
vv
v
v
v v
v
Skolai Arc
low TiO₂ Nikolailower group
high TiO₂ Nikolaiupper group
Maier and Barnes (2004) ascribe sub-chondri�c Pt:Pd ra�os to the refractory behavior of Pt-alloys under condi�ons of dry mel�ng associated with tholeii�c basalts; these phases are more fusible in the more fluid-rich mel�ng regimes of metasoma�zed lithospheric mantle. Alterna�vely, Hughes et al. (2015) suggest that elevated (near-chondrite) Pt:Pd ra�os in early magmas of the North Atlan�c Igneous Province are the result of assimila�on of fusible phases in previously Pt-enriched subcon�nental lithospheric mantle. That the PGE-enriched rocks of the upper suite (e.g., picrites) have the same Pt:Pd ra�o as the early, lower magma�c suite may suggest that these later magmas assimilated residual PGE-bearing phases from the lower magma conduit. Early segrega�on of a sulfide melt from the low TiO₂ magma could also explain the lack of PGE mineraliza�on associated with this suite. The PGE endowment of the low TiO₂ phase may have been assimilated in the upper crust by the subsequent high TiO₂ magma.
A) B)
DePaolo (1981). Trace element and isotopic effects of combined wallrock assimila�on and frac�onal crystalliza�on. Earth Planet. Sci. Le�ers, 53, 189–202.; Ghiorso, Hirschmann, Reiners, Kress, (2002). The pMELTS: A revision of MELTS for improved calcula�on of phase rela�ons and major element par��oning related to par�al mel�ng of the mantle to 3 GPa. G3, 3(5), 1–35.; Greene, Scoates, Weis, Israel, (2005). Flood basalts of the Wrangellia Terrane, southwest Yukon: Implica�ons for the forma�on of oceanic plateaus, con�nental crust and Ni-Cu PGE mineraliza�on: Yukon Explor. and Geol. 2004, 109–120.; Greene, A.R., Scoates, J.S., Weis, D., and Israel, S., 2005, Flood basalts of the Wrangellia Terrane, southwest Yukon: Implica�ons for the forma�on of oceanic plateaus, con�nental crust and Ni-Cu PGE mineraliza�on: Yukon Explora�on and GeologyGreene, Scoates, Weis, (2008). Wrangellia flood basalts in Alaska: A record of plume-lithosphere interac�on in a Late Triassic accreted oceanic plateau. G3, 9(December).; Greene, Scoates, Weis, Israel, (2009). Geochemistry of Triassic flood basalts from the Yukon (Canada) segment of the accreted Wrangellia oceanic plateau. Lithos, 110(1-4), 1–19.; Hughes, McDonald, Kerr, (2015). Pla�num-group element signatures in the North Atlan�c Igneous Province: Implica�ons for mantle controls on metal budgets during con�nental breakup. Lithos, 233, 89-110.; Maier, Barnes, (2004). Pt/Pd and Pd/Ir ra�os in mantle-derived magmas: A possible role for mantle metasoma�sm. S African J. Geol., 107, 333-340.; McDonough, Sun, (1995). The composi�on of the earth. Chem. Geol. 120, 223–253.; Pearce, (2008). Geochemical fingerprin�ng of oceanic basalts with applica�ons to ophiolite classifica�on and the search for Archean oceanic crust. Lithos, 100 (1), 14-48.; Pearce, Cann, (1973). Tectonic se�ng of basic volcanic rocks determined using trace element analyses. Earth Planet. Sci. Le�ers, 19.2, 290-300.; Nokleberg, Aleinikoff, Lange, Silva, Miyaoka, Schwabl, Zehnerl, (1992). Preliminary geologic map of the Mount Hayes Quadrangle, eastern Alaska Range, Alaska. U.S. Dept. of the Interior, U.S.G.S. Open-File Report 92-594.; Taylor, Premo, Meier, Taggart, (2008). The metallogeny of Late Triassic ri�ing of the Alexander terrane in southeastern Alaska and northwestern Bri�sh Columbia. Econ. Geol., 103, 89–115.; Twelker, Bachmann, Freeman, Newberry, Reioux, Sicard, Tuzzolino, Wright, Wypych, (2014). Major-Oxide, Minor-Oxide, and Trace-Element Geochemical Data from Rocks and Stream Sediments in the Wrangellia Mineral Assessment Area, Gulkana, Healy, Mount Hayes, and Talkeetna Mountains Quadrangles, Alaska. Div. of Geol. & Geoph. Surveys Raw-Data File 2014-3.; Workman, Hart, (2005). Major and trace element composi�on of the depleted MORB mantle (DMM). Earth Planet. Sci. Le�ers, 231(1-2), 53–72.; Wypych, Twelker, Freeman, Lande, Newberry, Reioux, Sicard, (2014). Major-Oxide and Trace-Element Geochemical Data from Rocks Collected in 2014 in the Wrangellia Mineral Assessment Area, Talkeetna Mountains C-4, C-3, and B-4 Quadrangles, Alaska. Div. of Geol. & Geoph. Surveys Raw-Data File 2014-22.; Wypych, Twelker, Freeman, Lande, Newberry, Reioux, Sicard, (2015). Major-Oxide and Trace-Element Geochemical Data From Rocks Collected in 2015 in the Wrangellia Mineral Assessment Area, Alaska. Div. of Geol. & Geoph. Surveys Raw-Data File 2015-16.
A) B)
maficand ultramaficintrusion
