1Brown University, 2University of California Santa Cruz, 3Monterey Bay Aquarium Research Institute, 4United States Geological SurveyPortner, R.A.1, Dreyer, B.D.2, Clague, D.A.3, Lowenstern, J.B.4, Head, J.W.1, Saal, A.E.1Degassing history of a mid-ocean ridge rhyolite dome on the Alarcón Rise, Gulf of CaliforniaV11C-4739
I: Introduction
V: Conclusions and highlights
ReferencesGualda, G.A.R., Ghiorso, M.S., Lemons, R.V., and Carley, T.L., 2012, J Pet, v. 53, p. 875-890Head, J.W., and Wilson, L., 2003, JVGR, v. 121, p. 155-193Lowenstern, J.B., and Pitcher, B.W., 2013, Amer Mnrlgst, v. 98, p. 1660-1668Newman, S., and Lowenstern, J.B., 2002, Comp Geosc, v. 28, p. 597-604Newman, S., Stolper, E.M., and Epstein, S. (1986) Amer Mnrlgst, 71, 1527–1541.Schneider, C.A., Rasband, W.S., and Eliceiri, K.W., 2012, Nat Meth, v. 9, p. 671-675Shea, T., Houghton, B.F., Gurioli, L., Cashman, K.V., Hammer, J.E., and Hobden, B.J., 2010, JVGR, v. 190, p. 271-289
Attenuated total reflectance (ATR) and transmission FTIR spectroscopy was used to measure volatile concentrations in lava glasses and glass inclusions (GI) in rhyolite phenocrysts. ATR experiments were standardized using a correction factor based on a best fit line through samples with known H2O concentrations (Fig. 5). Polished thin sections used for ATR experiments greatly reduces the time needed to prepare doubly polished wafers for GI analysis. H2O contents in Alarcon Rise volcanic rocks increase as expected from fractional crystallization. Total H2O in mineral host glasses increase from 0.15-0.25% in basalt to 1.4-2.3% in andesite, dacite and rhyolite. (Fig. 6). H2O contents within intermediate-felsic compositions agree well with estimated saturation limits at the depth of eruption (2480-2340 mbsl). CO2 was only detected (lower limit ~30 ppm) in basaltic compositions using transmission experiments (Fig. 6-inset). Glass inclusions (GI) from rhyolitic plagioclase (An 14-23) and olivine (Fo 8-10) contain 2.5-4.6% and 2.1-3.6% H2O, respectively (Fig. 7). CO2 was not detected in any rhyolitic phenocrysts using ATR (lower limit ~150 ppm). Embayment textures between olvine and host glass (Fig. 3B) were also analyzed and contain slightly higher H2O (2.5-4.5%) than smaller GI in the same crystals.
A 2325-2475 meter deep rhyolite lava dome and surrounding intermediate-mafic complex on the Alarcón Rise mid-ocean ridge in the Gulf of California was sampled extensively during a 2012 Monterey Bay Aquarium Research Institute expedition (Fig.1). The dome occurs near the northern end of the rise 8-5 km inboard from the Pescadero Transform. Whole rock and mineral geochemistry suggests that the dome formed entirely through oceanic crustal processes (see B. Dreyer et al.,V31B-4746 on Wednesday morning). Volcanic rock and volcaniclastic samples from on and around the dome were analysed for volatile (CO2 and H2O) contents using FTIR spectroscopy to better understand the degassing history of the dome and the formation of its pumiceous carapace.
Intermediate to mafic lava compositions are entriely pillow in form (Fig. 2A) whereas rhyolites exhibit tabular and breccia morphologies (Fig. 2B). Rhyolites are dense to sparsely vesicular (<10%), microlite-rich and contain a variety of phenocryst minerals (Fig. 3). Local deposits of pumiceous lapilli-ash also occur (Fig. 2C) and include dense to highly vesicular components (Fig. 2D). Pumiceous lapilli contain highly vesicular fracture networks that separate dense “pseudoclasts" (Fig. 2E-H). Textures and major element geochemistry suggest that both dernse and highly vesicular lithologies originated from the same magma that formed the majority of the dome (Fig. 4). Lapilli vesicularities of 40-67% were measured using bimodal corrected SEM images in ImageJ and FOAMS software protocols (Fig. 2I). Bulk vesicularities of pumiceous lapilli (Fig. 2E) range from 30-47% (using 2.28 g/cm3 dense rock equivelent).
Volcanic rocks and volcaniclastic deposits on the Alarcón Rise mid-ocean ridge preserves volatile enrichment and degassing during fractional crystallization from basalt to rhyolite. FTIR spectroscopy indicate that rhyolite phenocrysts crystallized 1-3 km deep in the crust and that their host melts exsolved a maximum of 3% H2O during ascent to the seafloor. This exsolved volatile phase translates to 53% vesicles upon eruption (after Head and Wilson 2003), and is in good agreement with highly vesiculated (30-68%) pumiceous lapilli. Although this threshold is slightly less than what would be required for magmatic fragmentation (3.6% exsolved H2O and 75% vesiculation), the presence of highly vesicular ash particles representing fragmented pumiceous breccia argues otherwise. Moreover, the localized high vesicularity of pumiceous deposits is in stark contrast to the sparsely vesicular (<10%) nature of lavas making up the majority of the rhyolite dome. We posit that decoupled volatiles from a deeper magma body migrated through fracture networks to the seafloor causing mild explosivity and ash dispersal.
Vesicularity =62% - ImageJ software47% - Foams software
Fracti
onal crysta
lizatio
n model
AcknowledgementsWork funded by NSF grant # OCE-1355436.We would like to thank the crew and ROV pilots of the RV Western Flyer for assistance with sample collection and ship time. J. Paduan, J. Martin, D. Caress, R. Spelz, N. Botto, and S. Roeske helped with bathymetric data processing, sample preparation, and data analysis.
pmc
bob
cob
II: Sample characteristics
Figure 2: Sample characteristics A) Andesite pillows on steep slope.
B) Tabular-fractured rhyolite on steep slope.
C) Rhyolitic pumiceous lapilli-ash/breccia. Sediment push core is 30 cm long.
D) Sparsley vesicular clear and brown obsidian (cob and bob), and pumiceous (pmc) ash shards.
E) Pumiceous lapilli with dense "pseudoclasts" (clear to brown) separated by highly vesicular fracture network.
F) Pseudoclasts (pc) with highly vesiculated margins that bound open fractures (frac). Plagioclase (plag) phenocrysts are fragmented.
G) Close up of pseudoclast (pc) margin showing coarse-, fine- and very fine-grained vesicles.
H) Very fine-grained (<1 µm) vesicular region.
I) Bimodal color image used for vesicularity calculations in ImageJ (Schneider et al. 2012) and Foams (Shea et al. 2010) software. The latter accounts for 3D stereological conversion.
BE
FD
pc
pc
pc
plagfra
c
pc
C
30 cm
10 mm
30 cm
A
I
GH
olvplag
glass
cpx
zrc
A
100 µm
Figure 3: Rhyolite mineralogy A) Cross/plane polarized light image of dense rhyolite glass with plagioclase (plag), olivine (olv) and clinopyroxene (cpx) phenocrysts with accessory zircon (zrc) locally. Ilmenite, titanomagnetite, pyrite and rare orthophyroxene also occur.
B) Olivine crystal with glass inclusions (GI) and amoeboid shape embayments.
B
emba
ymen
t
GI
olivine
100 µm
F
(
(
(
USA
Mexico
EPR
Fig. 1B
108°W
109°W
24°N
TT
PT
AR
(
Figure 1: Study area maps A) Location map of study area in the mouth of the Gulf of California. Red lines represent rift segments, green lines depict transforms and blue line indicates subduction zone (teeth on hangingwall). B) Seabeam 2112 bathymetric data (50 m cell size) of the Alarcon Rise (AR) and bounding Tamayo Transform (TT) and Pescadero Transform (PT).
C) AUV bathymetric map (1 m cell size ) of rhyolite dome and ROV sample collection. Lavas were collected by ROV manipulator arms and volcaniclastic samples were recovered with sediment push cores and scoop bags.
A B
Fig. 1C
23°N
Figure 4: Minerals and lava chemistry Major elements from phenocrysts (bottom) and their host glasses (top). Many host basalts and some basaltic andesites include samples from Alarcón Rise not shown on Fig. 2C. Error bars in bottom represent 1sigma deviation from mean mineral compositions. Dashed line represents fractional crystallization model calculated using 100 MPa pressure and QFM-1 (Gualda et al. 2012) .
III: Lava and mineral chemistry Lava compositions exhibit trends in major elements that are corroborated by trends trhough average olivine and plagioclase phenocryst compositions (Fig. 4). Increased deviation in mineral compositions of intermediate rocks require localized disequilibria and magma-mixing (see B. Dreyer et al. V31B-4746 on Wednesday).
RhyoliteDaciteAndesite
Basaltic andesite
0
10
20
30
40
50
60
70
80
90
Oliv
ine
Fo
80 1020304050607090
Basalt
Plagioclase An
Olivine and Plagioclase
ATR
expe
rimen
ts
Tran
smis
sion
ex
perim
ents
Rhyolite lavaDacite
AndesiteBasaltic andesite
2
4
6
8
45 50 55 60 65 70 75
Host glass Na2 O
+ K2 O
(wt%
)
Host glass SiO2 (wt%)
Basalt
Host glassRhyolite breccia
Basalt (B)
Basaltic andesite (BA)
Andesite (A)
Dacite (D) Rhyolite (R)Rhyolite dome
MBSL2300
24500 125 250 375 500
MetersC
Fig. 2B
Fig. 2C
volcaniclastic/sediment
Fig. 2A
MBSL0
4000100500Kilometers
IV: FTIR methods and results
Figure 5: FTIR methodology and ATR standarizationComparison of data collected by two different Germanium ATR attachments shows that each is unique and must be calibrated against known standards. ATR attachments must also be calibrated (focused) per manufacturer parameters. Accepted H2O contants in known rhyolite, dacite and basalt standards are based on transmission and manometry experiments (Lowenstern and Pitcher, 2013 and references therein). Insets show spectra from transmission and ATR experiments on the same Alarcon Rise rhyolite sample. The height (in abosorbance units) of the 3450 cm-1 peak (A3450) in ATR spectra and sample density (e.g. 2.28 g/cm3 for Alarcón Rise rhyolite) are input to the equation defined by the red line to calculate total H2O by ATR. Peak heights of the NIR peaks were used to calculate total H2O (Newman et al. 1986) by transmission.
50
55
60
65
70
75
0.0 0.5 1.0 1.5 2.0
SiO
2 (w
t%)
total H2O (wt%)
BasaltsBasaltic andesite
Andesite
Dacite
Rhyolite breccia
H2 O
saturation (rhyolite) at
eruption depth (24 MPa)
Figure 7: Alarcón Rise rhyolite phenocryst glass inclusion (GI) tH2O:Positive trend between the A3450 (total H2O) and A1630 (molecular H2O) peaks in mineral GI, and their rhyolite host glasses. H2O solubility isobars were calculated using a 780°C rhyolite magma and no CO2 (Newman and Lowenstern, 2002). Estimated crustal depths of entrapment (dashed lines) are calculated using a crustal density of 2900 kg/m3 and correct for hydrostatic pressure at the depth of eruption (24 MPa).
Figure 6: Alarcón Rise host glass total H2OMeasured H2O contents show more variation and tend to be slightly higher in transmission vs. ATR experiments. H2O saturation range is modeled for 780°C rhyolite melt with no CO2. Basalt (49% SiO2) volatile solubility (inset) is for a 1200°C melt. Dashed line follows model in Fig. 4.
0.000
0.005
0.010
0.015
0.020
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
A 163
0 pe
ak h
eigh
t
H2O total (from A3450)
Host glass
Plagioclase inclusions
Olivine inclusions
Olivine embayment
25 MPa
(35 m)
50 MPa
(914m)
75 MPa
(1793 m)
100 MPa
(2672 m)
125 MPa
(3551 m)
0
100
200
0 1 2
CO2
(ppm
)
H2O (wt%)
40 MPa30
20
10
Rhyolite lava
Lava BrecciaAlarcon Rise rhyolite
-0.002
-0.006
-0.010
-0.014
1500250035004500
ATR = 1.82% H2O
0
0.4
0.8
1.2
1.6
2
15002500350045005500
0.08
0.12
0.16 transmission1.81% H2O
0.001
0.003
0.005
0.007
0.009
0.011
0.013
0.015
0 1 2 3 4 5 6 7
ATR
A34
50/d
ensit
y
total H2O ( wt%) by transmission and manometry
Rhyolite (this study)
DaciteBasalt
Rhyolite (Lowenstern and Pitcher, 2013)
This study R² = 0.979((A3450/density)*501)-0.468
R² = 0.998
Lowenstern and Pich
er (2013)
NIR abso
rban
ceab
sorb
ance
wavenumbers (cm-1)
0.0104
see spectra to right
0.027
0.034
StandardsAlarcón Rise
wavenumbers (cm-1)
Fractional
crysta
llizatio
n
model
Fractional crystallization model