Journal of Petrology Advance Access originally published online on May 27, 2008
Journal of Petrology 2008 49(7):1297-1318; doi:10.1093/petrology/egn025
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Geochemical Variations during K
lauea's Pu‘u ‘
‘
Eruption Reveal a Fine-scale Mixture of Mantle Heterogeneities within the Hawaiian Plume
1Department of Geology and Geophysics, University of Hawai‘i, Honolulu, HI 96822, USA
2Department of Geological Sciences, San Diego State University, San Diego, CA 92182, USA
3Department of Geosciences, University of Massachusetts, Amherst, MA 01003, USA
4Research School of Earth Sciences, Australian National University, Canberra A.C.T. 0200, Australia
RECEIVED AUGUST 15, 2007; ACCEPTED APRIL 21, 2008
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONLong-term geochemical monitoring of lavas from the continuing 25-year-old Pu‘u ‘
‘
eruption allows us to probe the crustal and mantle magmatic processes beneath K
lauea volcano in unparalleled detail. Here we present new Pb, Sr, and Nd isotope ratios, major and trace element abundances, olivine compositions, and petrographic data for Pu‘u ‘‘ lavas erupted from 1998 to 2005. Olivine fractionation and accumulation are important crustal processes for the eruption, with minor clinopyroxene fractionation observed in the most recent lavas. Small, yet systematic variations in 87Sr/86Sr and incompatible trace element ratios, and MgO-normalized major element abundances document rapid changes in the parental magma composition delivered to Pu‘u ‘‘. Recent (1998–2003) lavas display a systematic temporal evolution towards an intermediate area between the compositional fields of historical Klauea and Mauna Loa lavas. At least three distinct mantle source components are required to explain the overall isotopic and chemical variability of Pu‘u ‘‘ lavas. Two of these source components observed in pre-1998 Pu‘u ‘‘ lavas have similar Pb, Sr, and Nd isotope ratios, although one underwent a recent (< 8 ka) small-degree melting event and became depleted in incompatible trace elements. This recently depleted component was an increasingly important source for lavas erupted between 1985 and 1998. The third component is a hybrid mixture of nearly equal portions of Klauea- and Mauna Loa-like mantle source compositions. It was progressively tapped in greater amounts from 1998 to 2003 and then subsequently decreased. The increasing importance of the hybrid source can be explained if melt pathways migrated from an area within Klauea's typical melting region (important for the 1985–1998 lavas) towards Mauna Loa, where a similar proportion of Klauea- and Mauna Loa-like mantle components might exist. The Pu‘u ‘‘ data suggest that Kea and Loa mantle components are distributed on a fine-scale within the Hawaiian plume, and both are present beneath Klauea volcano. Based on the geochemical and isotopic variations during the Pu‘u ‘‘ eruption, the estimated volume for Klauea and Mauna Loa compositional heterogeneities is < 10–35 km3. KEY WORDS: Hawaii; Kilauea; volcanoes; geochemistry; mantle heterogeneity
|
INTRODUCTION |
|---|
Time-series studies of geochemical variations during long-lived eruptions or eruptive sequences provide valuable insight into the magmatic processes within active volcanoes (e.g. Arenal, Costa Rica, Bolge et al., 2006; Etna, Italy, Rizzo et al., 2006
; Grímsvötn, Iceland, Sigmarsson et al., 1992; Parícutin, Mexico, Wilcox, 1954; McBirney et al., 1987; Piton de la Fournaise, Réunion, Vlastelic et al., 2005). These studies provide a detailed view of the crustal- and mantle-related magmatic processes occurring on short time scales (days to years). Klauea volcano, located on the island of Hawai‘i (Fig. 1), has erupted (
4·3 km3; Macdonald et al., 1983; Sutton et al., 2003) frequently during the past 200 years, making it an ideal location to investigate temporal variations in lava chemistry that are related to changes in mantle source compositions and melting conditions within the Hawaiian plume.
|
Previous seismic and petrological studies suggest that K
lauea magmas originate from partial melting at mantle depths >60–80 km within the upper Hawaiian plume (e.g. Eaton & Murata, 1960; Watson & McKenzie, 1991; Tilling & Dvorak, 1993). Rapid melt extraction from the mantle source region into chemically isolated channels (e.g. McKenzie, 1985; Williams & Gill, 1989) is probably the dominant melt transport mechanism beneath Klauea because it provides a relatively large amount of melt for sustained, high-volume eruptions such as Pu‘u ‘‘ (Pietruszka et al., 2006). After accumulating, these melts are thought to ascend through a primary conduit delivering magma to Klauea's shallow (2–6 km deep) magma reservoir, and may subsequently erupt in the summit caldera or feed the volcano's two rift zones (e.g. Tilling & Dvorak, 1993; Wright & Klein, 2006).
The Pu‘u ‘‘ eruption is the longest sustained (25+ years) and most voluminous (3 km3 erupted lava) historical eruption of Klauea volcano (Garcia et al., 2000; Heliker & Mattox, 2003). Pu‘u ‘‘ magmas are thought to partially bypass the summit reservoir (based on the rapid variations in incompatible trace element ratios for Pu‘u ‘‘ lavas compared with Klauea summit lavas) before intruding Klauea's East Rift Zone to feed a shallow (< 3 km depth) magma reservoir system beneath the Pu‘u ‘‘ cone (Garcia et al., 1996; Shamberger & Garcia, 2006). Since the start of the eruption in January 1983, there have been small but systematic variations in the Pb, Sr, and O isotope ratios and major and trace element abundances of Pu‘u ‘‘ lavas. These fluctuations provide an unprecedented opportunity to document the crustal processes (e.g. crystal fractionation and accumulation, and crustal assimilation) and mantle source and melting variations (e.g. mantle source heterogeneity, and melt production, extraction, and transportation) within the Hawaiian plume on a time scale of months to years (e.g. Garcia et al., 1989, 1992, 1996, 1998, 2000; Putirka, 1997; Thornber, 2001, 2003; Pietruszka et al., 2006).
Two different length scales of mantle heterogeneity within the Hawaiian plume have been recognized based on the distinct isotopic variations of single volcanoes. Large-scale heterogeneity has been proposed based on the persistent intershield geochemical differences of Hawaiian volcanoes over tens to hundreds of thousands of years (e.g. Frey & Rhodes, 1993; Chen et al., 1996), and the long-term differences in Pb isotope ratios (Tatsumoto, 1978; Abouchami et al., 2005) along two NW–SE-trending loci of volcanoes (Fig. 1): the northeastern ‘Kea’ trend (e.g. Klauea) and the southwestern ‘Loa’ trend (e.g. Mauna Loa). The Kea end member is defined by higher 206Pb/204Pb and 143Nd/144Nd and lower 87Sr/86Sr and is predominantly observed in lavas from Klauea, Mauna Kea, West Maui, and East Moloka‘i volcanoes (Stolper et al., 1996; DePaolo et al., 2001; Blichert-Toft et al., 2003; Eisele et al., 2003; Xu et al., 2007). The Loa end member is defined by higher 87Sr/86Sr and lower 143Nd/144Nd and 206Pb/204Pb and is mostly observed in lavas from Mauna Loa, Hual
lai, Lana‘i, Kaho‘olawe, West Moloka‘i and Ko‘olau volcanoes (Hauri, 1996; Lassiter & Hauri, 1998; Abouchami et al., 2005; Fekiacova et al., 2007).
Single volcanoes (e.g. Klauea and Mauna Loa) also record isotopic variations over shorter time scales (years to centuries) that are attributed to partial melting of small-scale compositional heterogeneities within the plume (e.g. Frey & Rhodes, 1993; Kurz et al., 1995; Rhodes & Hart, 1995; Pietruszka & Garcia, 1999a; Marske et al., 2007). Estimates for the size and shape of small-scale heterogeneities range from vertical streaks that are several tens to hundreds of kilometers long (e.g. Farnetani et al., 2002; Eisele et al., 2003; Abouchami et al., 2005) to heterogeneous blobs set in a compositionally distinct matrix (e.g. Frey & Rhodes, 1993; Rhodes & Hart, 1995; Blichert-Toft et al., 2003). A range of vertical length scales for these compositional heterogeneities from 6·5–160 km (Blichert-Toft et al., 2003) to 0·06–12 km (Kurz et al., 2004) have been estimated based on isotopic fluctuations recorded in 550–180 ka Mauna Kea lavas. In contrast, the presence of a pancake-shaped heterogeneity (>18 km wide and <5–10 km thick) has been inferred based on systematic Pb, Sr, and Nd isotopic fluctuations in young prehistoric (< 2·6 ka) Klauea and Mauna Loa lavas (Marske et al., 2007). Unlike these previous studies, which document the size of mantle heterogeneities on a scale of hundreds to thousands of years, the Pu‘u ‘‘ eruption offers an opportunity to probe the finer-scale compositional variations related to distinctive mantle sources on very short time scales (months to years). Here we provide a comprehensive petrological evaluation of the most recent (1998–2005) Pu‘u ‘‘ lavas using petrography, olivine and whole-rock chemistry, and Pb, Sr, and Nd isotope ratios, and discuss the compositional evolution of these lavas and their implications for the nature and scale of mantle source heterogeneity within the Hawaiian plume.
|
OVERVIEW OF THE PU‘U ‘‘ ERUPTION
|
|---|
The onset of the eruption (episode 1) in January 1983, began with intermittent fire fountaining along an 8-km-long fissure system in the middle of K
lauea's East Rift Zone (Wolfe et al., 1987; Garcia et al., 1989; Fig. 1). During episodes 2 and 3, activity was localized to a 1 km section of the fissure system. A central vent, Pu‘u ‘‘, was the focus of effusion for episodes 4–47 (June 1983–June 1986). These episodes were generally short-lived (5–100 h) with variable (10–400 m) lava fountaining heights (Garcia et al., 1992; Heliker & Mattox, 2003). In July 1986, the primary vent migrated 3 km downrift from the Pu‘u ‘‘ cone to the K
paianaha lava shield (Fig. 1). This shift coincided with a change in eruptive style from episodic, fire-fountaining events to nearly continuous and gentle effusion (Garcia et al., 1996). Kpaianaha was the site of nearly continuous lava effusion (episodes 48 and 49) until February 1992, when activity shifted back to Pu‘u ‘‘. From February 1992 to January 1997 (episodes 50–53), a shield 60 m high and 1· 3 km in diameter was built at Pu‘u ‘‘ (Heliker et al., 1998). On January 29, 1997, the lava lake inside the Pu‘u ‘‘ shield suddenly drained, and a 22 h eruption (episode 54) occurred 2–4 km uprift (Harris et al., 1997; Thornber et al., 2003).
Following a 24 day hiatus, episode 55 began, marking the longest (1997–2007) and most voluminous (1· 6 km3) effusive interval for this eruption. Episode 55 activity displayed nearly continuous eruption of lava (except for infrequent 1–4 day pauses) from vents on the south and west flanks of the Pu‘u ‘‘ cone (Garcia et al., 2000; Heliker & Mattox, 2003). The most significant event during episode 55 occurred on September 12, 1999, when magma-induced earthquake swarms and surficial deflation of the Pu‘u ‘‘ cone were followed by intrusion of magma into the upper East Rift Zone of Klauea (Nakata et al., 2000). An 11 day hiatus followed as magma supplying the Pu‘u ‘‘ cone was temporarily diverted to the upper rift zone. Episode 55 ended after a 1 day eruption (episode 56) that occurred 6 km uprift of the Pu‘u ‘‘ cone in mid-June 2007 (Poland et al., 2008). Following a 2 week pause, lava effusion resumed in early July 2007, and, as of May 2008, lava continues to erupt from a vent 2 km downrift of the Pu‘u ‘‘ cone.
|
PETROGRAPHY |
|---|
A majority of samples studied here were collected in a molten state and most were quenched in water. The sample label (e.g. 17-Aug-01) is the date it was collected, and in nearly all cases is within a day of when it was erupted. The vast majority of the 1998–2005 Pu‘u ‘
‘ lavas petrographically studied are glassy, strongly vesicular, friable and aphyric to moderately olivine-phyric (< 3 vol. % phenocrysts; Table 1). Olivine is almost always the only phenocryst in these samples and is usually small (0·5–1 mm in diameter), euhedral and undeformed with spinel and glass inclusions. Olivine is somewhat less common (1 vol. %) in the 1998–2005 lavas compared with those from the preceding 6 years of eruptive activity but similar in abundance to lavas erupted from the Kpaianaha vent (1986–1992; Garcia et al., 1996). The total olivine abundance (phenocrysts and microphenocrysts) generally correlates with whole-rock MgO content, although it can vary 4 vol. % for a given MgO (Table 1). Clinopyroxene phenocrysts are rare in the 1998–2005 Pu‘u ‘‘ lavas, yet microphenocrysts of clinopyroxene (up to 3 vol. %) are present in almost all of these lavas (Table 1). In contrast, clinopyroxene is absent or rare (<1 vol. %) in earlier Pu‘u ‘‘ lavas (except in the more evolved 1983 lavas; Garcia et al., 1992, 1996). Clinopyroxene crystals are small (0·1–0·3 mm), occur commonly in clusters of 3–12 grains, and commonly display sector zoning. Plagioclase microphenocrysts occur in most of the 1998–2005 Pu‘u ‘‘ lavas, although they are usually rare (< 1 vol. %; Table 1) and small (0·1–0·2 mm wide laths). Plagioclase is less common in earlier erupted Pu‘u ‘‘ lavas (except for the lavas from episodes 1–10 and 54, which were affected by magma mixing in Klauea's East Rift Zone (Garcia et al., 1989, 1992, 2000; Thornber et al., 2003). The groundmass of the 1998–2005 lavas generally consists of honey–brown glass or black cryptocrystalline material with microlites (<0·1 mm) of plagioclase, clinopyroxene, olivine, and spinel.
|
|
OLIVINE COMPOSITION |
|---|
A five spectrometer Cameca SX-50 electron microprobe with SAMx automation was used for the olivine analyses at the University of Hawai‘i using techniques described by Garcia et al. (2000
). Olivine compositions (Table 2) were determined for 175 olivine crystals from 19 lavas erupted between 1998 and 2005 that span a wide compositional range (whole-rock MgO contents of 6·7–9·5 wt %). All of the analyzed olivine crystals are unzoned or normally zoned with up to 3% forsterite (Fo) variation from core to rim. The forsterite content of the olivine cores range from 76·5 to 86·0% (Fig. 2) with phenocrysts and microphenocrysts overlapping in composition. The average Fo content is 81·1%, which is similar to lavas from the previous 6 years and somewhat lower than olivines in the earlier lavas (1986–1992) erupted from the Kpaianaha vent (82·5% on average; Garcia et al., 1996, 2000). The NiO and CaO contents in the olivines are moderate (Table 2) indicating crystallization at crustal depths from somewhat fractionated parental magmas (e.g. Garcia, 2002).
|
|
Olivines in the 1998–2005 lavas typically have Fo compositions that are in equilibrium with their whole-rock Mg-number, particularly for lavas with lower Mg-number (<58; Fig. 2). Most samples with Mg-number >58 have olivine compositions that plot below the equilibrium field, especially the sample with the highest Mg-number (4-Jan-00; Fig. 2). The higher Mg-number lavas probably accumulated olivine, which is consistent with their higher abundance of this mineral (e.g. 4-Jan-00 contains the highest olivine content; Table 1). The highest measured forsterite content of olivine within the 1998–2005 lavas (Fo86) occurs in a sample with an intermediate Mg-number, 21-Jan-98 (Table 2). This olivine, like all of the other crystals, shows no signs of deformation. Thus, it is probably indicative of the parental magma composition for recent Pu‘u ‘
‘ lavas (Mg-number 59; Fig. 2), consistent with estimates for previous Pu‘u ‘‘ lavas (Garcia et al., 2000). |
WHOLE-ROCK ANALYTICAL METHODS |
|---|
Eighty-two new Pu‘u ‘
‘ lava samples erupted between 1998 and 2005 were analyzed for major and trace element (Rb, Sr, Y, Nb, Zr, Zn, Ni, Cr, V, Ba, and Ce) abundances over a 7 year period using X-ray fluorescence spectrometry (XRF) at the University of Massachusetts (Table 3). The Klauea basaltic standards collected from the same flow, K1919 (n = 13) and BHVO-1 (n = 13), were run as internal controls for major and trace element abundances during this period, respectively (Table 3). All Pu‘u ‘‘ samples presented here were analyzed in the same XRF laboratory using the same calibration procedures. Thus, any long-term analytical drift in the major and trace element abundances is expected to be relatively minor. Details of methods used and estimates of analytical precision for the XRF analyses have been given by Rhodes (1996) and Rhodes & Vollinger (2004).
|
In addition, 27 samples were analyzed over a 1 week period for Sc, V, Cr, Co, Ni, Cu, Zn, Sr, Cs, Rb, Ba, Th, U, Nb, Zr, Hf, Y, and rare earth element (REE) abundances using inductively coupled plasma mass spectrometry (ICP-MS) at the Australian National University (Table 4). The K
lauea rock standard BHVO-2 (n = 2) was analyzed with these samples. Analytical methods and estimates of precision for the ICP-MS trace element analyses are given in Table 4 and by Norman et al. (1998). Prior to both XRF and ICP-MS analyses, the Pu‘u ‘‘ lava samples were washed in an ultrasonic bath of deionized water for 10–20 min, hand picked (30–100 g) to remove any rare altered rock chips, and powdered in a tungsten carbide swing mill for the XRF analyses and an agate mill for the ICP-MS analyses.
|
Fourteen Pu‘u ‘
‘ lavas erupted between 1998 and 2005 were analyzed for Pb and Nd isotope ratios (Table 5) by multi-collector (MC)-ICP-MS using a Nu Plasma system at San Diego State University (SDSU). Strontium isotope ratios were measured using this instrument and/or by thermal ionization mass spectrometry (TIMS) using a VG Sector 54 instrument at SDSU to compare the results of the two instruments. Additionally, four lavas erupted from 1989 to 1998 (Garcia et al., 1992, 1996, 2000) were reanalyzed for Sr isotope ratios from the original dissolutions using TIMS to improve the analytical precision for these samples (Appendix, Table A1). A detailed overview of the analytical methods used in this study has been given by Marske et al. (2007). Additional details pertinent to this study are presented in Table 5.
|
|
GEOCHEMISTRY OF 1998–2005 PU‘U ‘‘ LAVAS
|
|---|
The 1998–2005 Pu‘u ‘
‘ lavas are compositionally similar to earlier lavas from this eruption (Fig. 3). For example, their MgO contents (6·7–9·8 wt %) lie within the range of 1985–1998 lavas (6·7–10·1 wt %, excluding mixed and evolved lavas from episode 54). However, small, but significant short-term (years) variations in major element abundances (at a given MgO) are evident in these lavas (Fig. 3). Similar variations have been observed for historical Klauea summit lavas and were related to changes in parental magma composition (e.g. Wright, 1971; Garcia et al., 2003). Overall, Pu‘u ‘‘ lavas have become progressively lower in CaO/Al2O3 ratios and incompatible element (TiO2 and K2O) contents, and higher in SiO2 abundances (at a given MgO content) during the eruption (even if more differentiated samples with <7·2 wt % MgO are excluded; Fig. 3). At a given MgO content, the total variation of 
(i.e. total iron), Al2O3, and Na2O abundances (not shown) for the 1998–2005 lavas lie within the compositional field of previous lavas. The high SiO2 and low CaO, TiO2, and K2O contents, and CaO/Al2O3 ratios of the recent lavas expand the known compositional range for historical Klauea lavas towards historical Mauna Loa lavas (Fig. 3).
|
error bars are shown in the corner of each plot unless they are smaller than the size of the symbols.The relatively low abundances and ratios of incompatible elements (e.g. Nb and La/Sm) in 1998–2005 lavas form trends that partially overlap with 1989–1997 Pu‘u ‘
‘ lavas, but also expand the compositional range to the lowest values observed during the eruption (Fig. 4). Recent Pu‘u ‘‘ lavas continue the overall temporal decrease in highly to moderately incompatible trace element ratios (i.e. La/Yb) and abundances (i.e. Nb and Ba) since the early part of the eruption (Figs 4 and 5). These trace element abundances and ratios of the recent lavas, like the major element contents, extend the Pu‘u ‘‘ compositional range toward the field of historical Mauna Loa lavas (Fig. 4).
|
|
The 87Sr/86Sr ratios of the 1998–2005 lavas (0·70360–0·70364) are higher than those of previous Pu‘u ‘
‘ lavas (0·70357–0·70360). In contrast, Pb (Figs 5 and 6) and Nd isotope ratios are within the range of previous lavas. However, the 1998–2005 lavas make a small, yet distinctive trend in 206Pb/204Pb vs 87Sr/86Sr that expands the isotopic range for this eruption (Fig. 6). Unlike the major and trace element chemistry, these Pb and Sr isotope variations trend towards an area between the compositional fields of Klauea and Mauna Loa (rather than directly towards the field of Mauna Loa lavas). Furthermore, this trend does not project towards a Hawaiian mantle end member (e.g. Kea or Loa) or towards the known isotopic composition of any other Hawaiian volcano.
|
|
CRUSTAL MAGMATIC PROCESSES DURING THE ERUPTION |
|---|
Olivine fractionation plays a dominant role in controlling the compositional variations in Hawaiian lavas (e.g. Powers, 1955
; Wright, 1971). The importance of olivine crystallization in Pu‘u ‘‘ lavas is evident from their wide range of MgO contents (5·6–10·1 wt %), and the presence of normally zoned olivine phenocrysts (Garcia et al., 1996, 2000). Shallow magma mixing between stored rift magmas and ‘fresh’ MgO-rich magma(s) was an important process that controlled the composition of lavas erupted before February 1985 (episodes 1–29; Garcia et al., 1992), and during episode 54 in January 1997 (Garcia et al., 2000; Thornber et al., 2003). Most (93%) Pu‘u ‘‘ lavas (excluding evolved lavas from episodes 1–29 and 54) have >7·2 wt % MgO, suggesting that the variation in the major element abundances of these lavas is primarily related to olivine fractionation and/or accumulation (i.e. olivine control; Wright, 1971). However, the MgO and Ni abundances of the Pu‘u ‘‘ lavas have systematically decreased (with much scatter) since the eruption location shifted in early 1992 from the Kpaianaha vent to the Pu‘u ‘‘ cone (Figs 3 and 5), suggesting that Pu‘u ‘‘ magmas are becoming increasingly differentiated with time. Clinopyroxene fractionation has become an increasingly important process for 1998–2005 lavas. For example, there is a greater abundance of clinopyroxene microphenocrysts in the 1998–2005 lavas (up to 3 vol. %; Table 1) compared with earlier erupted Pu‘u ‘‘ lavas (<1 vol. % or absent), and 50% of the most recent lavas (2004–2005) have differentiated beyond olivine control (<7·2 wt % MgO; Fig. 5).
Olivine accumulation has also affected some Pu‘u ‘‘ lava compositions based on the low Fo contents of some olivines compared with their whole-rock Mg-number (Fig. 2). For example, the maximum MgO difference for two olivine-controlled 1998–2005 lavas (23-Jul-04, 7·0 wt % MgO; 4-Jan-00, 9·5 wt % MgO) with similar olivine Fo contents (81–82%; Fig. 2) can be explained by the accumulation of 6·1 vol. % olivine. This is consistent with the 5% greater modal abundance of olivine phenocrysts in sample 4-Jan-00 compared with 23-Jul-04, and the position of sample 4-Jan-00 to the right of the equilibrium field in Fig. 2. All of the 2003–2005 lavas analyzed have Fo contents in equilibrium with their bulk compositions (Fig. 2).
|
TEMPORAL COMPOSITIONAL VARIATIONS |
|---|
Pu‘u ‘
‘ lavas display systematic temporal variations of MgO-normalized major element abundances, ratios of highly to moderately incompatible trace elements (e.g. La/Yb), highly incompatible trace element ratios (e.g. Ba/Nb), and Pb and Sr isotope ratios (Fig. 5). At least three distinct end-member magma compositions may be delineated based on correlated temporal changes among some MgO-normalized major element abundances, and incompatible trace element and Sr isotope ratios: (1) 1985–1998 (days 760–5500); (2) 1998–2003 (days 5501–7400); (3) 2003–2005 (days 7401–8400). Magma mixing affected the composition of lavas erupted before February 1985 (days 1–745; Garcia et al., 1992) and during episode 54 in January 1997 (day 5141; Garcia et al., 2000; Thornber et al., 2003), and these lavas were excluded from the plots.
The SiO2 temporal trend for Pu‘u ‘‘ lavas inversely correlates with the temporal variations of CaO and TiO2 (Fig. 5). Between 1985 and 1998, MgO-normalized major element abundances display relatively flat (SiO2) and slightly decreasing (CaO and TiO2) trends. However, lavas erupted between 1998 and 2003 display changes in slope with significant increases and decreases in these abundances, respectively, until mid-2003 (Fig. 5). Following a compositional reversal in mid-2003, the MgO-normalized SiO2 abundances have decreased whereas CaO and TiO2 contents have increased. In contrast, Al2O3 and Fe2O3 contents (not shown) have remained nearly constant.
Pu‘u ‘‘ lavas also display systematic fluctuations in incompatible trace element ratios that correlate with the major element changes (Fig. 5). Between 1985 and 1998 there are small but significant decreases in some ratios of incompatible trace elements until day 5500. Between 1998 and 2003 these incompatible trace element ratios display flattening (e.g. La/Yb) or nearly constant (e.g. Ba/Nb) temporal trends. Some ratios of highly over moderately incompatible trace elements (e.g. La/Yb) record a small compositional reversal in mid-2003, followed by a small increase similar to the reversal recorded by the normalized major element abundances (Fig. 5).
The 87Sr/86Sr ratios of Pu‘u ‘‘ lavas gradually increased between 1985 and 1998, before a sharper increase occurred in 1998 (day 5500). This shift coincided with the significant increases in MgO-normalized SiO2 abundances and decreases in the CaO and TiO2 abundances (Fig. 5). The 87Sr/86Sr ratios increased to the highest observed values for this eruption in mid-2003, and reversed from 2003 to 2005. This is also analogous to the reversals of MgO-normalized major element abundances and incompatible trace element ratios (Fig. 5). In contrast, the Pb and Nd isotope ratios have remained relatively constant since 1985 (relative to analytical error), especially for the 1998–2005 lavas (Fig. 5; Table 5).
In summary, three distinct end-member compositions are important during the Pu‘u ‘‘ eruption. First, an early end member (1985) has relatively low 87Sr/86Sr ratios and MgO-normalized SiO2 abundances, and high CaO and TiO2 abundances and incompatible trace element ratios (e.g. Ba/Nb or La/Yb). Second, a later end member (1998) from a recently depleted source (Pietruszka et al., 2006) has slightly higher 87Sr/86Sr and MgO-normalized SiO2 contents, yet lower incompatible trace element ratios and CaO and TiO2 abundances. Third, the most recent end-member composition (2003) displays the highest 87Sr/86Sr ratios and MgO-normalized SiO2 abundances, and lowest abundances of CaO and TiO2 and ratios of incompatible trace elements for the eruption.
|
PU‘U ‘‘ SOURCE CHARACTERISTICS
|
|---|
A lithospheric source component for Pu‘u ‘
‘ lavas?Magmas originating from partial melting within the Hawaiian plume can be compositionally modified at shallower depths by the assimilation of hydrothermally altered oceanic crust (e.g. Eiler et al., 1996
) or lower gabbroic crust (e.g. Gaffney et al., 2004), or by partial melting of the upper ambient mantle (lithosphere or asthenosphere) beneath Hawai‘i (e.g. Tatsumoto, 1978; Chen & Frey, 1985; Stille et al., 1986; Lassiter et al., 1996). However, it is unlikely that crustal assimilation or melting of the upper ambient mantle significantly modified the chemical signature of Pu‘u ‘‘ lavas. Pu‘u ‘‘ lavas erupted between 1983 and 1986 have relatively low 
18O groundmass values (4·6–5·0
) that are in disequilibrium with their olivine phenocrysts, suggesting that these early magmas interacted with shallow wall rock in the rift zone just prior to eruption (Garcia et al., 1998). The switch in eruptive style from episodic lava fountaining to near-continuous effusion in 1986 led to a marked reduction or elimination of contamination, based on higher 18O groundmass values (5·0–5·3) in equilibrium with olivines in the 1986–1998 lavas (Garcia et al., 1998). Furthermore, the 1998–2003 Pu‘u ‘‘ lavas record systematic temporal increases in 87Sr/86Sr ratios and relatively constant 206Pb/204Pb ratios (Fig. 6) that trend away from the compositional fields of lithospheric mantle xenoliths from Salt Lake Crater, O‘ahu (Okano & Tatsumoto, 1996) and Huallai volcano, Hawai‘i (Lassiter & Hauri, 1998), and Cretaceous Pacific mid-oceanic ridge basalts near Hawai‘i (e.g. Ocean Drilling Program Site 843; King et al., 1993; Fekiacova et al., 2007).
Historical Klauea summit lavas have relatively low 18O isotope values that are attributed to <5–12% contamination of parental magmas with altered country rock from both Klauea and Mauna Loa (Garcia et al., 2008). The overall increase of Sr isotopes in Pu‘u ‘‘ lavas could potentially be explained if a more typical Klauea parental magma (e.g. 1993–1997 Pu‘u ‘‘ lavas) progressively assimilated a roughly constant (55:45) mixture of older Klauea and Mauna Loa basement rocks (Fig. 6). However, this would require 100% contamination because the Pu‘u ‘‘ lava with the highest 87Sr/86Sr ratio overlaps with the composition of the 55:45 Klauea–Mauna Loa assimilant. These combined observations suggest that melt interaction with the upper mantle, crust, or volcanic edifice beneath Klauea is minimal for recent Pu‘u ‘‘ lavas.
A third mantle source for the Pu‘u ‘‘ eruption
At least three distinct mantle source components are required to explain the compositional variability of Pu‘u ‘‘ lavas (Fig. 7). The 1985–1998 Pu‘u ‘‘ lavas originated from at least two distinct source components with similar Klauea-like Pb, Sr, and Nd isotopic compositions but different incompatible element abundances and ratios (Garcia et al., 2000). One component with higher MgO-normalized CaO and TiO2 abundances and incompatible trace element ratios (e.g. Ba/Ce or La/Yb) was important during the early part of the eruption (1985; Figs 5 and 7), following the period of magma mixing during episodes 1–29. The temporal decreases among 230Th–238U and 226Ra–230Th disequilibria, incompatible trace element ratios (e.g. Th/U or Nd/Sm), and some normalized major element abundances in lavas from 1985 to 2001 (Fig. 5) suggest that a second mantle component was tapped (Pietruszka et al., 2006). This ‘recently depleted’ component is thought to have formed at <8 ka (based on modeling of 226Ra–230Th–238U disequilibria) by the removal of melt from Klauea's source region within the Hawaiian plume, causing it to become depleted in incompatible trace elements (Pietruszka et al., 2006). A progressive increase in the proportion of the recently depleted component is indicated by the temporal trends between 1985 and 1998 (Figs 5 and 7; Garcia et al., 2000; Pietruszka et al., 2006).
|
The systematic geochemical variations from 1998 to 2003 (Figs 5–7
) require a third component that was increasingly tapped during this time interval. Based on the temporal increases of some highly incompatible trace element ratios (e.g. Ba/Th) in lavas erupted from 1999 to 2001, Pietruszka et al. (2006) suggested that Pu‘u ‘‘ lavas are derived in greater proportions of a source component similar to historical Mauna Loa lavas. However, simple mixing of melt from a Mauna Loa-like source (with relatively low 206Pb/204Pb and high 87Sr/86Sr) with an earlier Pu‘u ‘‘ composition (i.e. the recently depleted component) cannot explain the increase in 87Sr/86Sr ratios at relatively constant 206Pb/204Pb ratios between 1998 and 2003 (Fig. 6).
Rhodes et al. (1989) proposed that magma from Mauna Loa may periodically invade Klauea's plumbing system. Pre-mixing of nearly equal proportions of historical Klauea- and Mauna Loa-like magmas prior to eruption could potentially explain the trends of recent Pu‘u ‘‘ lavas (Figs 6 and 8). Although this pre-mixing could occur in Klauea's 2–3 km3 summit reservoir (Pietruszka & Garcia, 1999b), the rapid compositional changes in Pu‘u ‘‘ lavas are inconsistent with mixing in this reservoir (Garcia et al., 1996). No other suitable crustal reservoir is known to accommodate this magma mixing. Thus, crustal magma mixing is an unlikely explanation for the 1998–2003 Pu‘u ‘‘ compositional variations. Instead, these recent lavas can be explained if they were derived from a mixture of Mauna Loa- and Klauea-like mantle sources that subsequently melted. This hybrid source represents a new component for the eruption.
|
Historical lavas from K
lauea and Mauna Loa volcano provide an important window to the present-day composition and distribution of mantle components in the Hawaiian plume (e.g. Rhodes & Hart, 1995; Pietruszka & Garcia, 1999a). Consequently, the origin of the recent Pu‘u ‘‘ lavas is discussed below in terms of mixing between the mantle source components defined by historical lavas from these two volcanoes (rather than the more extreme end-member isotopic compositions observed in Ko‘olau and Mauna Kea lavas). The proportions of Klauea and Mauna Loa components in the hybrid source can be estimated from compositional mixing trends (Figs 6 and 8). The 1917 Klauea lava was chosen as an end member because it has the highest 206Pb/204Pb and 143Nd/144Nd and lowest 87Sr/86Sr ratios among olivine-controlled historical Klauea lavas (Pietruszka & Garcia, 1999a). Although a range of historical Mauna Loa lavas would make reasonable isotopic end members for this calculation, the 1887 Mauna Loa lava (Rhodes & Hart, 1995) was selected because it creates a suitable mixing trend on the plots of MgO-normalized major elements vs 206Pb/204Pb and 87Sr/86Sr (Fig. 8). The 10-Jan-1997 Pu‘u ‘‘ sample was chosen to represent the recently depleted Klauea source component because it has the lowest ratios of highly incompatible trace elements (e.g. Ba/Nb or Ba/Rb) for this eruption. Mixing trends between the 1917 Klauea (55%) and 1887 Mauna Loa (45%) lavas (Figs 6 and 8) pass within analytical error of the 2001–2003 Pu‘u ‘‘ lavas (i.e. the samples with the highest 87Sr/86Sr and MgO-normalized SiO2 values). Therefore, this 55:45 Klauea–Mauna Loa composition might be a good estimate of the hybrid source.
This mixing model suggests that the melt contribution from the recently depleted source component decreased starting in early 1998 as melt derived from the hybrid source was tapped in greater proportions until mid-2003 (Figs 7 and 8). Following the mid-2003 compositional reversal, the lavas display chemical and isotopic variations that overlap with the compositional fields of the 1998–2003 lavas (Figs 3–8), indicating a diminishing importance for the hybrid component since 2003.
A pyroxenite source for Pu‘u ‘‘ lavas?
Partial melting of a heterogeneous plume source containing a mixture of peridotite and ancient recycled oceanic crust ± sediment (pyroxenite or eclogite) has become a common explanation for the chemical and isotopic variations in Hawaiian lavas (e.g. Hauri, 1996; Lassiter & Hauri, 1998; Blichert-Toft et al., 1999; Takahashi & Nakajima, 2002; Gaffney et al., 2005; Sobolev et al., 2005, 2007; Herzberg, 2006). For example, Ko‘olau lavas, with relatively high 87Sr/86Sr (0·7044) and SiO2 (53–55 wt %), are explained by melting ancient recycled oceanic crust within the Hawaiian plume (Hauri, 1996; Lassiter & Hauri, 1998; Blichert-Toft et al., 1999; Huang & Frey, 2005; Fekiacova et al., 2007). Further support for a pyroxenite source within the Hawaiian plume comes from modeling compositional variations of lavas during long-lived eruptions, including Pu‘u ‘‘ (Reiners, 2002). This model predicts that the continuous SiO2 increases and CaO decreases could be explained if Pu‘u ‘‘ lavas originated from a mixed pyroxenite–peridotite source with different solidi.
The temporal increases in 87Sr/86Sr ratios and SiO2 abundances (normalized to 10 wt % MgO) in the 1998–2003 lavas (Fig. 5) could be evidence for increased melting of an eclogite or pyroxenite lithology (i.e. recycled oceanic crust) in the Hawaiian plume. However, an increasing contribution of this source lithology during the eruption is unlikely for the following reasons. (1) The MgO-normalized SiO2 trend is relatively flat prior to 1998, increased from 1998 to 2003, and has decreased since mid-2003. Moreover, the CaO trend has increased since mid-2003. Both trends are inconsistent with a simple mixed lithology source (e.g. Reiners, 2002). (2) The long-term decreases in CaO abundances (normalized to 10 wt % MgO; Fig. 5) from 1985 to 2003 suggest that there has been a decrease in the amount of clinopyroxene that is melted in the mantle source region, rather than the predicted increase of this mineral. (3) The Ni abundances of the lavas have progressively decreased (at a given MgO) from 1992 to 2005 (Fig. 3). Because Ni is highly compatible in olivine relative to clinopyroxene (e.g. Sobolev et al., 2005), the decreases in Ni content suggest that recent Pu‘u ‘‘ lavas originated from a peridotite source that became more olivine-rich and/or clinopyroxene-poor with time. A peridotite source is also supported by the positive correlation between 226Ra–230Th and 230Th–238U disequilibria of 1985–2001 Pu‘u ‘‘ lavas (Pietruszka et al., 2006).
|
SMALL-SCALE MANTLE HETEROGENEITY |
|---|
The timing of the temporal inflections of the 87Sr/86Sr ratios in the Pu‘u ‘
‘ lavas may be used to help constrain the scale of heterogeneity within Klauea's melting region. The steady increase in the 87Sr/86Sr ratios between 1998 (day 5500) and 2003 (day 7400) suggests that the proportion of the hybrid component progressively increased during this period. The lack of significant changes in the volume of magma stored in the shallow summit reservoir beneath Klauea during prolonged (months to years) historical Klauea rift eruptions (i.e. Mauna Ulu and Pu‘u ‘‘) suggests that the magma supply rate is similar to lava effusion rate (Tilling et al., 1987; Dvorak & Dzurisin, 1993; Denlinger, 1997). Assuming that the magma supply rate is approximately equal to the lava effusion rate (0·13 km3/year; Sutton et al., 2003) for the Pu‘u ‘‘ eruption, the total volume of melt extracted from Klauea's source region from 1998 to 2003 (1900 days) was 0·7 km3. This estimate probably represents the maximum volume of melt derived from the hybrid source during this period. If 100% of the recently depleted and hybrid components were being tapped at the temporal inflections in 1998 and 2003, respectively, melt from the hybrid source might represent 50% of the total lava volume erupted from 1998 to 2003 (0·35 km3). Models for tholeiitic basalt production within the Hawaiian plume suggest that the melt-zone porosity within Klauea's source region is 2–3% (Sims et al., 1999; Pietruszka et al., 2001). Thus, if melt tapped from 1998 to 2003 represents 2–3% of the total volume from which it was extracted, then the volume of the source region that supplied melt during this period would be 10–35 km3.
The 1998–2003 lavas plot within analytical error along the mixing line from the recently depleted source (10-Jan-1997) towards the hybrid component (Figs 6 and 8). If these recent lavas are derived from a hybrid mantle source containing a mixture of Klauea and Mauna Loa components, then the size of these components must be significantly smaller than the volume of the source region that was tapped from 1998 to 2003 (<10–35 km3). Thus, the Klauea and Mauna Loa mantle components that make up the hybrid component are thought to be mixed on a fine scale in the Hawaiian plume.
|
CHEMICAL STRUCTURE OF THE HAWAIIAN PLUME |
|---|
The long-term geographical and compositional differences between Hawaiian shield volcanoes have been related to the distribution of large-scale heterogeneities in a radially (Lassiter et al., 1996
; Bryce et al., 2005), asymmetric (Abouchami et al., 2005) or irregularly zoned (Kurz et al., 2004) Hawaiian plume. For example, Klauea and Mauna Loa volcanoes have erupted geochemically distinct lavas for most of their known eruptive history, suggesting that the source components of these volcanoes (i.e. Kea and Loa) have remained compositionally distinct [except for some young prehistoric (AD 900–1400) Klauea and Mauna Loa lavas; Marske et al., 2007] over a time scale of thousands of years (e.g. large-scale heterogeneity; Frey & Rhodes, 1993; Abouchami et al., 2005). However, the recent Pu‘u ‘‘ compositional trends suggest that both Klauea- and Mauna Loa-like components are present within Klauea's source region with a spatial distribution capable of creating the rapid fluctuation towards the hybrid composition on a time scale of years.
To explain the recent trend of Pu‘u ‘‘ lavas towards the hybrid composition we propose a model with large-scale compositional heterogeneity (e.g. Lassiter et al., 1996; DePaolo et al., 2001; Bryce et al., 2005) that is gradational across the Hawaiian plume (black to white shading in Fig. 9a). On a finer scale, Klauea- and Mauna Loa-like heterogeneities (<10–35 km3) are assumed to be present (within the enlarged circles in Fig. 9a), but vary in relative abundance depending on location within the Hawaiian plume. For example, the darker zonation within Klauea's typical source region contains more Klauea heterogeneities, and vice versa for the whiter zone below Mauna Loa. Similarly, the intermediate gray zone located between these two volcanoes represents a source with approximately equal amounts of Klauea- and Mauna Loa-like compositions (Fig. 9a).
|
Resolving the spatial distribution of these small-scale compositional heterogeneities (<10–35 km3) is problematic given the wide range in estimates for the size of K
lauea's melting region. These estimates vary from an 55 km thick region near the central axis of the plume (Watson & McKenzie, 1991) to a thickness of <5–10 km (Marske et al., 2007). The maximum radius of this melting region is probably 17 km (i.e. half the distance between the summits of Klauea and Mauna Loa; Pietruszka et al., 2001). Estimates for the rates of mantle melting in the Hawaiian plume range from >0·0005 to >0·03 kg/m3 per year (Cohen et al., 1993; Hemond et al., 1994; Sims et al., 1999; Pietruszka et al., 2001). Even the highest melting rates would require melting over a voluminous mantle source region (8500 km3; Pietruszka et al., 2006) to account for the vigorous lava effusion rate during the Pu‘u ‘‘ eruption (0·13 km3/year; Sutton et al., 2003). Tapping such a large melting region would probably homogenize the melts derived from the three distinct sources for this eruption. Instead, the compositional variability of Pu‘u ‘‘ lavas could be preserved if compositionally distinct melts are extracted into chemically isolated channels and efficiently transported to the surface (Pietruszka et al., 2006).
The melting region beneath Hawaiian volcanoes is predicted to be zoned, with higher degrees of partial melting (and higher melt productivity) in a relatively thin zone near the top of the melting region (Watson & McKenzie, 1991). Thus, melt extraction would probably be more effective if it occurred laterally over a thinner (i.e. 5–10 km thick; Marske et al., 2007) region. As melt migrates into channels to supply the Pu‘u ‘‘ eruption, it must be extracted from more distal areas to sustain the flow of melt to the surface, otherwise the melt supply would become exhausted (Pietruszka et al., 2006). In this context, we propose that the systematic geochemical trends toward the hybrid composition from 1998 to 2003 could be explained if melt pathways migrated from an area within Klauea's typical melting region dominated by the early and recently depleted component (black ovals; Fig. 9b) towards Mauna Loa, where more Mauna Loa-like components would be expected (Fig. 9c). In contrast, the MgO-normalized SiO2 abundances are thought to be controlled by the depth of partial melting (e.g. Hirose & Kushiro, 1993; Kushiro, 1996; Longhi, 2002). Thus, the temporal increase in normalized SiO2 abundances in 1998–2003 lavas could also be explained if melt production and segregation occurred at progressively shallower depths (e.g. Stolper et al., 2004) during this interval (Fig. 9b and c). This model for a fine-scale mixture of compositionally distinct mantle heterogeneities (i.e. Klauea and Mauna Loa components) within the Hawaiian plume is consistent with the presence of both Kea and Loa compositions in young prehistoric (AD 900–1400) Klauea and Mauna Loa lavas (Marske et al., 2007), and in the melt inclusions of East Maui lavas (Ren et al., 2006).
|
CONCLUSIONS |
|---|
The Pu‘u ‘
‘ eruption is exceptional among historical eruptions for its long duration (25+ years) and compositional variability. The systematic geochemical fluctuations in Pu‘u ‘‘ lavas document the short-term crustal (e.g. crystal fractionation) and mantle (melting and source heterogeneity) processes in the Hawaiian plume. Pu‘u ‘‘ lavas erupted from 1985 to 1998 are thought to have originated from at least two distinct source components with similar isotopic compositions, although one was more depleted in incompatible trace elements by a recent (<8 ka) melting event in the Hawaiian plume. Post-1998 Pu‘u ‘‘ lavas record small but distinctive variations of MgO-normalized major element abundances, and Sr isotope and incompatible trace element ratios (compared with earlier erupted lavas) that require a third source component. Lavas erupted between 1998 and 2003 display a temporal geochemical evolution toward an intermediate area between the compositional fields of historical Klauea and Mauna Loa lavas. Based on mixing models, the 1998–2003 Pu‘u ‘‘ lavas trend towards a hybrid mantle source composition made of roughly equal proportions of Klauea- and Mauna Loa-like components. The contribution from a recently depleted Klauea component decreased starting in early 1998 as the volcano tapped greater proportions of a hybrid component until mid-2003. The systematic geochemical trends toward this hybrid composition can be explained if melt pathways migrated from an area within Klauea's melting region (important for 1985–1998 lavas) towards Mauna Loa, where an equal mixture of Klauea- and Mauna Loa-like components may exist. The presence of Klauea (i.e. Kea) and Mauna Loa (i.e. Loa) components (<10–35 km3) in Pu‘u ‘‘ lavas suggests that both of these components are present as a fine-scale mixture in Klauea's source region. |
APPENDIX |
|---|
|
|
ACKNOWLEDGEMENTS |
|---|
We thank the numerous workers who collected samples for this study (especially Virginia Aragon, Kate Bridges, Eric Haskins, Bruce Houghton, Laslo Keszthely, and Scott Rowland), Chad Shishado and Nicole Robinson for assistance with the rock preparation, Michael Vollinger for the XRF preparation work and analyses, and Joan Willis for her support in the isotope clean lab at SDSU. We thank Amy Gaffney, Karen Harpp, and Mark Kurz for their helpful reviews. This research was supported by two grants from the National Science Foundation to M.O.G. and A.J.P. (EAR 03-36874 and 07-38817). This paper is SOEST Contribution 7159.
*Corresponding author. Telephone: + 1-808-956-5960. E-mail: marske{at}hawaii.edu
|
REFERENCES |
|---|
Abouchami W, Hofmann AW, Galer SJG, Frey FA, Eisele J, Feigenson M. Lead isotopes reveal bilateral asymmetry and vertical continuity in the Hawaiian mantle plume. Nature (2005) 434:851–856.[CrossRef][Medline]
Blichert-Toft J, Albarède F, Frey F. Hf isotopic evidence for pelagic sediments in the source of Hawaiian basalts. Science (1999) 285:879–882.
Blichert-Toft J, Weis D, Maerschalk C. Hawaiian hotspot dynamics as inferred from the Hf and Pb isotopic evolution of Mauna Kea volcano. Geochemistry, Geophysics, Geosystems (2003) 4. doi:10.1029/2002GC000340.
Bolge LL, Carr MJ, Feigenson MD, Alvarado GE. Geochemical stratigraphy and magmatic evolution at Arenal volcano, Costa Rica. Journal of Volcanology and Geothermal Research (2006) 157:34–48.[CrossRef][Web of Science]
Bryce JG, DePaolo DJ, Lassiter JC. Geochemical structure of the Hawaiian plume: Sr, Nd, and Os isotopes in the 2·8 km HSDP-2 section of Mauna Kea volcano. Geochemistry, Geophysics, Geosystems (2005) 6. doi:10.1029/2004GC000809.
Byers CD, Garcia MO, Muenow DW. Volatiles in pillow rim glasses from Loihi and Kilauea volcanoes, Hawaii. Geochimica et Cosmochimica Acta (1985) 49:1887–1896.[CrossRef][Web of Science]
Chen C-Y, Frey FA. Trace element and isotopic geochemistry of lavas from Haleakala volcano, East Maui, Hawaii: implications for the origin of Hawaiian basalts. Journal of Geophysical Research (1985) 90:8743–8768.
Chen C-Y, Frey FA, Rhodes JM, Eaton RM. Temporal geochemical evolution of Kilauea volcano: comparison of Hilina and Puna Basalt. Earth Processes: Reading the isotopic code, American Geophysical Union Geophysical Monograph—Basu A, Hart S, eds. (1996) 95:161–181.
Cohen AS, O’Nions RK. Melting rates beneath Hawaii: evidence from uranium series isotopes in recent lavas. Earth and Planetary Science Letters (1993) 120:169–175.[CrossRef][Web of Science]
Denlinger RP. A dynamic balance between magma supply and eruption rate at Kilauea volcano, Hawaii. Journal of Geophysical Research (1997) 102:18091–18100.[CrossRef]
DePaolo DJ, Bryce JG, Dodson A, Shuster DL, Kennedy BM. Isotopic evolution of Mauna Loa and the chemical structure of the Hawaiian plume. Geochemistry, Geophysics, Geosystems (2001) 3. doi:10.1029/2000GC000139.
Dvorak JJ, Dzurisin D. Variations in magma supply rate at Kilauea volcano, Hawaii. Journal of Geophysical Research (1993) 98:22255–22268.[CrossRef]
Eaton JP, Murata KJ. How volcanoes grow [Hawaii]. Science (1960) 132:925–938.
Eiler JM, Farley KA, Valley JW, Hofmann AW, Stolper EM. Oxygen isotope constraints on the sources of Hawaiian volcanism. Earth and Planetary Science Letters (1996) 144:453–468.[CrossRef][Web of Science]
Eisele J, Abouchami W, Galer SJG, Hofmann AW. The 320 kyr Pb isotope evolution of Mauna Kea lavas recorded in the HSDP-2 drill core. Geochemistry, Geophysics, Geosystems (2003) 4. doi:10.1029/2002GC000339.
Farnetani CG, Legras B, Tackley PJ. Mixing and deformation in mantle plumes. Earth and Planetary Science Letters (2002) 196:1–15.[CrossRef][Web of Science]
Fekiacova A, Abouchami W, Galer SJG, Garcia MO, Hofmann AW. Origin and temporal evolution of Koolau volcano, Hawaii: inferences from isotope data on the Koolau Scientific Drilling Project (KSDP), the Honolulu Volcanics and ODP Site 843. Earth and Planetary Science Letters (2007) 261:65–83.[CrossRef][Web of Science]
Frey FA, Rhodes JM. Intershield geochemical differences among Hawaiian volcanoes: implications for source compositions, melting process and magma ascent paths. Philosophical Transactions of the Royal Society of London (1993) 342:121–136.[Web of Science][Medline]
Gaffney AM, Nelson BK, Blichert-Toft J. Geochemical constraints on the role of oceanic lithosphere in intra-volcano heterogeneity at West Maui, Hawaii. Journal of Petrology (2004) 45:1663–1687.
Gaffney AM, Nelson BK, Blichert-Toft J. Melting in the Hawaiian plume at 1–2 Ma as recorded at Maui Nui: The role of eclogite, peridotite, and source mixing. Geochemistry, Geophysics, Geosystems (2005) 6. doi:10.1029/2005GC000927.
Galer SJG, Abouchami W. Practical application of lead triple spiking for correction of instrumental mass discrimination. Mineralogical Magazine (1998) 62A:491–492.[CrossRef]
Garcia MO. Submarine picritic basalts from Koolau volcano, Hawaii: implications for parental magma composition and mantle source. Hawaiian Volcanoes: Deep Underwater Perspectives. Geophysical Monographs, American Geophysical Union—Takahashi E, Lipman PW, Garcia MO, Naka J, Aramaki S, eds. (2002) 128:391–401.
Garcia MO, Ho RA, Rhodes JM, Wolfe EW. Petrologic constraints on rift zone processes: results from episode 1 of the Puu Oo eruption of Kilauea volcano, Hawaii. Bulletin of Volcanology (1989) 52:81–96.[CrossRef][Web of Science]
Garcia MO, Rhodes JM, Ho RA, Ulrich G, Wolfe EW. Petrology of lavas from episodes 2–47 of the Puu Oo eruption of Kilauea volcano, Hawaii: evaluation of magmatic processes. Bulletin of Volcanology (1992) 55:1–16.[CrossRef][Web of Science]
Garcia MO, Rhodes JM, Trusdell FA, Pietruszka AP. Petrology of lavas from the Puu Oo eruption of Kilauea volcano: III. The Kupaianaha episode (1986–1992). Bulletin of Volcanology (1996) 58:359–379.[CrossRef][Web of Science]
Garcia MO, Ito E, Eiler J, Pietruszka A. Crystal contamination of Kilauea volcano magmas revealed by oxygen isotope analysis of glass and olivine from the Puu Oo eruption lavas. Journal of Petrology (1998) 39:803–817.[CrossRef][Web of Science]
Garcia MO, Pietruszka AJ, Rhodes JM, Swanson K. Magmatic processes during the prolonged Puu Oo eruption of Kilauea volcano, Hawaii. Journal of Petrology (2000) 41:967–990.
Garcia MO, Pietruszka AJ, Rhodes JM. A petrologic perspective of Kilauea volcano's summit magma reservoir. Journal of Petrology (2003) 44:2313–2339.
Garcia MO, Ito E, Eiler JM. Oxygen isotope evidence for chemical interaction of Kilauea historical magmas with basement rocks. Journal of Petrology (2008) 49:757–769.
Harris AJL, Keszthelyi L, Flynn LP, Mouginis-Mark PJ, Thornber C, Kauahikaua J, Sherrod D, Trusdell F, Sawyer MW, Flament P. Chronology of the episode 54 eruption at Kilauea volcano, Hawaii, from GOES-9 satellite data. Geophysical Research Letters (1997) 24:3281–3284.[CrossRef][Web of Science]
Hauri EH. Major-element variability in the Hawaiian mantle plume. Nature (1996) 382:415–419.[CrossRef]
Heliker C, Mattox TN. The first two decades of the Puu Oo–Kupaianaha eruption; chronology and selected bibliography. In: Heliker, C. Swanson, D. A. & Takahashi, T. J. (eds). The Puu Oo–Kupaianaha Eruption of Kilauea Volcano, Hawaii: The First 20 Years. US Geological Survey Professional Papers (2003) 1676:121–136.
Heliker C, Mangan MT, Mattox TN, Kauahikaua JP, Helz RT. The character of long-term eruptions: inferences from episodes 50–53 of the Puu Oo–Kupaianaha eruption of Kilauea volcano. Bulletin of Volcanology (1998) 59:381–393.[CrossRef][Web of Science]
Hemond C, Hofmann AW, Heusser G, Condomines M, Raczek I, Rhodes JM. U–Th–Ra systematics in Kilauea and Mauna Loa tholeiites. Chemical Geology (1994) 116:163–180.[CrossRef][Web of Science]
Herzberg C. Petrology and thermal structure of the Hawaiian plume: a view from Mauna Kea. Nature (2006) 444:605–609.[CrossRef][Medline]
Hirose K, Kushiro I. Partial melting of dry peridotites at high pressure: determination of composition of melts segregated from peridotite using aggregates of diamonds. Earth and Planetary Science Letters (1993) 114:477–489.[CrossRef][Web of Science]
Huang S, Frey FA. Recycled oceanic crust in the Hawaiian plume: evidence from temporal geochemical variations within the Koolau Shield. Contributions to Mineralogy and Petrology (2005) 149:556–575.[CrossRef][Web of Science]
King AJ, Waggoner DG, Garcia MO. Geochemistry and petrology of basalts from Leg 136, central Pacific Ocean. In: Proceedings of the Ocean Drilling Program, Scientific Results, 136.—Wilkens RH, Firth J, Bender J, et al, eds. (1993) College Station, TX: Ocean Drilling Program. 107–118.
Kurz MD, Kammer DP. Isotopic evolution of Mauna Loa volcano. Earth and Planetary Science Letters (1991) 103:257–269.[CrossRef][Web of Science]
Kurz MD, Kenna TC, Kammer DP, Rhodes JM, Garcia MO. Isotopic evolution of Mauna Loa Volcano: a view from the submarine southwest rift zone. Mauna Loa Revealed: Structure, Composition, History and Hazards, American Geophysical Union Geophysical Monograph—Rhodes JM, Lockwood JP, eds. (1995) 92:289–306.
Kurz MD, Curtice J, Lott DE III, Solow A. Rapid helium isotopic variability in Mauna Kea shield lavas from the Hawaiian Scientific Drilling Project. Geochemistry, Geophysics, Geosystems (2004) 5. doi:10.1029/2002GC000439.
Kushiro I. Partial melting of fertile mantle peridotite at high pressures: an experimental study using aggregates of diamond. Earth Processes: Reading the isotopic code, American Geophysical Union Geophysical Monograph—Basu A, Hart S, eds. (1996) 95:109–122.
Lassiter JC, Hauri EH. Osmium-isotope variations in Hawaiian lavas: evidence for recycled oceanic lithosphere in the Hawaiian plume. Earth and Planetary Science Letters (1998) 164:483–496.[CrossRef][Web of Science]
Lassiter JC, DePaolo DJ, Tatsumoto M. Isotopic evolution of Mauna Kea volcano: results from the initial phase of the Hawaii Scientific Drilling Project. Journal of Geophysical Research (1996) 101:11769–11780.[CrossRef]
Longhi J. Some phase equilibria systematics of lherzolite melting: I. Geochemistry, Geophysics, Geosystems (2002) 2. doi:10.1029/2001GC000204.
Macdonald GA, Abbott AT, Peterson FL. Volcanoes in the Sea: the Geology of Hawaii. (1983) Honolulu, HI: University of Hawaii Press.
Marske JP, Pietruszka AJ, Weis D, Garcia MO, Rhodes JM. Rapid passage of a small-scale heterogeneity through the melting regions of Kilauea and Mauna Loa volcanoes, Hawaii. Earth and Planetary Science Letters (2007) 259:34–50.[CrossRef][Web of Science]
McBirney AR, Taylor HP, Armstrong RL. Paricutin re-examined; a classic example of crustal assimilation in calc-alkaline magma. Contributions to Mineralogy and Petrology (1987) 95:4–20.[CrossRef][Web of Science]
McKenzie D. 230Th–238U disequilibrium and the melting processes beneath ridge axes. Earth and Planetary Science Letters (1985) 72:149–157.[CrossRef][Web of Science]
Mittelstaedt E, Garcia MO. Modeling the sharp compositional interface in the Puu Oo magma reservoir, Kilauea volcano, Hawaii. Geochemistry, Geophysics, Geosystems (2007) 8. doi:10.1029/2006GC001519.
Moore JG, Ault WU. Historic littoral cones in Hawaii. Pacific Science (1965) 19:3–11.
Nakata JS, Heliker C, & Sherrod D. Hawaiian Volcano Observatory; Summary 99; Part 1, Seismic data, January to December 1999, with a chronological summary. US Geological Survey Report (2000) 61. OF 00-0433.
Norman MD, Griffin WL, Pearson NJ, Garcia MO, O’Reilly SY. Quantitative analysis of trace element abundances in glasses and minerals: a comparison of laser ablation ICPMS, solution ICPMS, proton microprobe, and electron microprobe data. Journal of Analytical Atomic Spectrometry (1998) 13:477–482.[CrossRef][Web of Science]
Okano O, Tatsumoto M. Petrogenesis of ultramafic xenoliths from Hawaii inferred from Sr, Nd, and Pb isotopes. Earth Processes: Reading the isotopic code, American Geophysical Union Geophysical Monograph—Basu A, Hart S, eds. (1996) 95:135–147.
Pietruszka AP, Garcia MO. A rapid fluctuation in the mantle source and melting history of Kilauea volcano inferred from the geochemistry of it historical summit lavas (1790–1982). Journal of Petrology (1999a) 40:1321–1342.[CrossRef][Web of Science]
Pietruszka AP, Garcia MO. The size and shape of Kilauea volcano's summit magma storage reservoir: a geochemical probe. Earth and Planetary Science Letters (1999b) 167:311–320.[CrossRef][Web of Science]
Pietruszka AP, Rubin KH, Garcia MO. 226Ra–230Th–238U disequilibria of historical Kilauea lavas (1790–1982) and the dynamics of mantle melting within the Hawaiian plume. Earth and Planetary Science Letters (2001) 186:15–31.[CrossRef][Web of Science]
Pietruszka AJ, Hauri EH, Carlson RW, Garcia MO. Remelting of recently depleted mantle within the Hawaiian plume inferred from the 226Ra–230Th–238U disequilibria of Puu Oo eruption lavas. Earth and Planetary Science Letters (2006) 244:155–169.[CrossRef][Web of Science]
Poland M, Miklius A, Orr T, Sutton J, Thornber D, Wilson D. New episodes of volcanism at Kilauea volcano, Hawaii. EOS Transactions, American Geophysical Union (2008) 89:37–38.[CrossRef]
Powers H. Composition and origin of basaltic magma of the Hawaiian Islands. Geochimica et Cosmochimica Acta (1955) 7:77–107.[CrossRef][Web of Science]
Putirka K. Magma transport at Hawaii: inferences based on igneous thermobarometry. Geology (1997) 25:69–72.
Reiners PW. Temporal–compositional trends in intraplate basalt eruptions: implications for mantle heterogeneity and melting processes. Geochemistry, Geophysics, Geosystems (2002) 3. doi:10.1029/2001GC000250.
Ren Z-Y, Tomoyuki S, Masako Y, Johnson KM, Takahashi E. Isotope compositions of submarine Hana Ridge lavas, Haleakala volcano, Hawaii: Implications for source compositions, melting process and the structure of the Hawaiian plume. Journal of Petrology (2006) 47:255–275.
Rhodes JM. Geochemical stratigraphy of lava flows sampled by the Hawaii Scientific Drilling Project. Journal of Geophysical Research (1996) 101:11729–11746.[CrossRef]
Rhodes JM, Hart SR. Episodic trace element and isotopic variations in historical Mauna Loa lavas: implications for magma and plume dynamics. Mauna Loa Revealed: Structure, Composition, History and Hazards, American Geophysical Union Geophysical Monograph—Rhodes JM, Lockwood JP, eds. (1995) 92:263–288.
Rhodes JM, Vollinger MJ. Composition of basaltic lavas sampled by phase-2 of the Hawaii Scientific Drilling Project: Geochemical stratigraphy and magma types. Geochemistry, Geophysics, Geosystems (2004) 5. doi:10.1029/2002GC000434.
Rhodes JM, Vollinger MJ. Ferric/ferrous ratios in 1984 Mauna Loa lavas: a contribution to understanding the oxidation state of Hawaiian magmas. Contributions to Mineralogy and Petrology (2005) 149:666–674.[CrossRef][Web of Science]
Rhodes JM, Wenz KP, Neal CA, Sparks JW, Lockwood JP. Geochemical evidence for invasion of Kilauea's plumbing system by Mauna Loa magma. Nature (1989) 337:257–260.[CrossRef]
Rizzo A, Caracausi A, Favara R, Martelli M, Paonita A, Paternoster M, Nuccio PM, Rosciglione A. New insights into magma dynamics during the last two eruptions of Mount Etna as inferred by geochemical monitoring from 2002 to 2005. Geochemistry, Geophysics, Geosystems (2006) 7. doi:10.1029/2005GC001175.
Roeder PL, Emslie RF. Olivine–liquid equilibrium. Contributions to Mineralogy and Petrology (1970) 29:275–289.[CrossRef][Web of Science]
Shamberger PJ, Garcia MO. Geochemical modeling of magma mixing and magma reservoir volumes during early episodes of Kilauea volcano's Puu Oo eruption. Bulletin of Volcanology (2006) 69:345–352.[CrossRef][Web of Science]
Sigmarsson O, Karlsson HR, Larsen G. The 1996 and 1998 subglacial eruptions beneath the Vatnajokull ice sheet in Iceland: contrasting geochemical and geophysical inferences on magma migration. Bulletin of Volcanology (2000) 61:468–76.[CrossRef][Web of Science]
Sims KWW, DePaolo DJ, Murrell MT, Baldridge WS, Goldstein S, Clague D, Jull M. Porosity of the melting zone and variations in the solid mantle upwelling rate beneath Hawaii: inferences from the 238U–230Th–226Ra and 235U–231Pa disequilibria. Geochimica et Cosmochimica Acta (1999) 63:4119–4138.[CrossRef][Web of Science]
Sobolev AV, Hofmann AW, Sobolev SV, Nikogosian IK. An olivine-free mantle source of Hawaiian shield basalts. Nature (2005) 434:590–597.[CrossRef][Medline]
Sobolev AV, Hofmann AW, Kuzmin DV, Yaxley GM, Arndt NT, Chung S-L, Danyushevsky LV, Elliott T, Frey FA, Garcia MO, Gurenko AA, Kamenetsky VS, Kerr AC, Krivolutskaya NA, Matvienkov VV, Nikogosian IK, Rocholl A, Sigurdsson IA, Sushchevskaya NM, Teklay M. The amount of recycled crust in sources of mantle-derived melts. Science (2007) 316:412–417.
Stille P, Unruh DM, Tatsumoto M. Pb, Sr, Nd, and Hf isotopic constraints on the origin of Hawaiian basalts and evidence for a unique mantle source. Geochimica et Cosmochimica Acta (1986) 50:2303–2319.[CrossRef][Web of Science]
Stolper EM, DePaolo DJ, Thomas DM. Introduction to special section; Hawaii Scientific Drilling Project. Journal of Geophysical Research (1996) 101:11593–11598.[CrossRef]
Stolper EM, Sherman S, Garcia MO, Baker M, Seaman C. Glasses in the submarine section of the HSDP-2 drill core, Hilo, Hawaii. Geochemistry, Geophysics, Geosystems (2004) 5. doi:10.1029/2003GC000553.
Sutton JA, Elias T. Lava-effusion rates for the Puu Oo–Kupaianaha eruption derived from SO2 emissions and very low frequency (VLF) measurements. The Puu Oo–Kupaianaha Eruption of Klauea Volcano, Hawaii: the First 20 Years. US Geological Survey Professional Papers—Heliker C, Swanson DA, Takahashi TJ, eds. (2003) 1676:121–136.
Takahashi E, Nakajima K. Melting process in the Hawaiian plume; an experimental study. Hawaiian Volcanoes: Deep Underwater Perspectives. American Geophysical Union, Geophysical Monographs—Takahashi E, Lipman PW, Garcia MO, Naka J, Aramaki S, eds. (2002) 128:403–418.
Tatsumoto M. Isotopic composition of lead in oceanic basalt and its implication to mantle evolution. Earth and Planetary Science Letters (1978) 38:63–87.[CrossRef][Web of Science]
Thirlwall MF. Multicollector ICP-MS analysis of Pb isotopes using a 207Pb–204Pb double spike demonstrates up to 400 ppm/amu systematic errors in Tl-normalization. Chemical Geology (2002) 184:255–274.[CrossRef][Web of Science]
Thornber CR. Olivine–liquid relations of lava erupted by Kilauea volcano from 1994 to 1998: implications for shallow magmatic processes associated with the ongoing East-Rift-Zone eruption. Canadian Mineralogist (2001) 39:239–266.[CrossRef][Web of Science]
Thornber CR. Magma-reservoir processes revealed by geochemistry of the Puu Oo–Kupaianaha eruption. In: Heliker, C. Swanson, D. A. & Takahashi, T. J. (eds). The Puu Oo–Kupaianaha Eruption of Kilauea Volcano, Hawaii: the First 20 Years. US Geological Survey Professional Papers (2003) 1676:121–136.
Thornber CR, Heliker C, Sherrod DR, Kauahikaua JP, Miklius A, Okubo PG, Trusdell FA, Budahn JR, Ridley WI, Meeker GP. Kilauea East Rift Zone magmatism; an episode 54 perspective. Journal of Petrology (2003) 44:1525–1559.
Tilling RI, Dvorak JJ. Anatomy of a basaltic volcano. Nature (1993) 363:125–133.[CrossRef]
Tilling RI, Wright TL, Millard HT Jr. Trace-element chemistry of Kilauea and Mauna Loa lava in space and time: a reconnaissance. Volcanism in Hawaii, US Geological Survey Professional Paper—Decker RW, Wright TL, Stauffer PH, eds. (1987) 1350:641–689.
Ulmer P. The dependence of the Fe2+–Mg cation-partitioning between olivine and basaltic liquid on pressure, temperature and composition. Contributions to Mineralogy and Petrology (1989) 101:261–271.[CrossRef][Web of Science]
Vlastelic I, Staudacher T, Semet M. Rapid change of lava composition from 1998 to 2002 at Piton de la Fournaise (Reunion) inferred from Pb isotopes and trace elements; evidence for variable crustal contamination. Journal of Petrology (2005) 46:79–107.
Wanless VD, Garcia MO, Trusdell FA, Rhodes JM, Norman MD, Weis D, Fornari DJ, Kurz MD, Guillou H. Submarine radial vents on Mauna Loa volcano, Hawaii. Geochemistry, Geophysics, Geosystems (2006) 7. doi:10.1029/2005GC001086.
Watson S, McKenzie D. Melt generation by plumes: a study of Hawaiian volcanism. Journal of Petrology (1991) 32:501–537.
West HB, Leeman WP. Isotopic evolution of lavas from Haleakala volcano, Hawaii. Earth and Planetary Science Letters (1987) 84:211–225.[CrossRef][Web of Science]
Wilcox RE. Petrology of Paricutin volcano, Mexico. US Geological Survey Bulletin (1954) B 0965-C:281–354.
Williams RW, Gill JB. Effects of partial melting on the uranium decay series. Geochimica et Cosmochimica Acta (1989) 53:1607–1619.[CrossRef][Web of Science]
Wolfe EW, Garcia MO, Jackson DB, Koyanagi RY, Neal CA, Okamura AT. The Puu Oo eruption of Kilauea volcano, episodes 1 through 20, January 3, 1983, to June 8, 1984. Volcanism in Hawaii, United States Geological Survey Professional Paper—Decker RW, Wright TL, Stauffer PH, eds. (1987) 1350:471–508.
Wright TL. Chemistry of Kilauea and Mauna Loa in space and time. United States Geological Survey Professional Paper (1971) 735:1–40.
Wright TL, Klein FW. Deep magma transport at Kilauea Volcano, Hawaii. Lithos (2006) 87:50–79.[CrossRef][Web of Science]
Xu G, Frey FA, Clague DA, Abouchami W, Blichert-Toft J, Cousens B, Weisler M. Geochemical characteristics of West Molokai shield- and postshield-stage lavas: constraints on Hawaiian plume models. Geochemistry, Geophysics, Geosystems (2007) 8. doi:10.1029/2006GC001554.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||