Ka'ena Volcano—A Precursor Volcano of the Island of O'ahu, Hawai'i

Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano—A precursor volcano of the island of O‘ahu, Hawai‘i

John M. Sinton1,†, Deborah E. Eason1, Mary Tardona1, Douglas Pyle1,§, Iris van der Zander1,#, Hervé Guillou2, David A. Clague3, and John J. Mahoney1,* 1University of Hawai‘i, Department of Geology and Geophysics, 1680 East-West Road, Honolulu, Hawaii 96822, USA 2Laboratoire des Sciences du Climat et de L’Environment, Commissariat à l’Energie Atomique et aux Energies Alternatives/Centre National de la Recherche Scientifi que, Gif sur Yvette, 91198, France 3Monterey Bay Aquarium Research Institute, Moss Landing, California 95039, USA

ABSTRACT O‘ahu, which we call Ka‘ena Volcano. After island of O‘ahu, and Ka‘ena Ridge appears to emergence, Ka‘ena Volcano tilted ~2° to the form a general prolongation of that volcano. Ka‘ena and Wai‘alu Ridges form promi- south. We estimate a total volume of 20–27 × However, Moore et al. (1989) alluded to the nent submarine ridges NW of the island of 103 km3 for Ka‘ena Volcano, taking into ac- possibility of a “third volcano forming Ka‘ena O‘ahu, Hawai‘i. We evaluate whether or count overlapping geometry of concurrently Ridge” that might have contributed debris to not either one of these ridges represents a active volcanoes. Sample compositions from the Wai‘anae slump, and Smith (2002) drew submarine extension of Wai‘anae Volcano the Ka‘ena landslide deposit are entirely con- attention to several enigmatic aspects of its on O‘ahu using new bottom observations, sistent with derivation from Ka‘ena, whereas morphology, including its large size in the geophysical surveys, and geochemical data most samples from the Wai‘anae slump are middle of the Kaua‘i Channel, and the pres- acquired on new samples from the region. likely derived from Wai‘anae Volcano. Uni- ence of “cone-shaped topographic highs” on it. Wai‘alu Ridge has the morphology of a sub- formly oriented dikes in the Wai‘anae NW Although Coombs et al. (2004) left the relation marine rift zone but is too shallow for its rift zone likely refl ect buttressing by a pre- of Ka‘ena Ridge to Wai‘anae unresolved, their distance from the O‘ahu shoreline; Ka‘ena exist ing Ka‘ena Volcano. Unusual isotopic fi gure 1 refl ects the possibility that it might Ridge also is unusually shallow and is sur- compositions of some Wai‘anae samples, in- be an extension of one of Wai‘anae Volcano’s mounted by two topographic shields. Ka‘ena cluding unique hydrous silicic lavas, probably rift zones, and Robinson and Eakins (2006) and Wai‘alu Ridges have similar magmatic refl ect interaction with underlying Ka‘ena included Ka‘ena Ridge in their calculation of and volcanic evolutionary histories, begin- crust. A newly recognized lava fl ow fi eld on the volume for Wai‘anae Volcano. ning ca. 5 Ma with a submarine, shield phase the southern fl ank of Ka‘ena Ridge extends The morphology of Ka‘ena Ridge differs of volcanism that produced high-SiO2, low- the previously known distribution of sec- from that of most Hawaiian rift zones, which FeO* tholeiites with higher 208Pb/204Pb than in ondary volcanism in the Kaua‘i Channel. steadily increase in depth with distance from the the adjacent Wai‘anae Volcano. Late-shield Putative submarine volcanic activity in the shoreline (Fig. 2) and have a hummocky appear- volcanism included transitional and alkalic region in 1956 cannot have built a large edi- ance refl ecting constitution predominantly of rock types, with lower SiO2 and enrichment fi ce and is unlikely to have produced pumice pillow lavas (Smith, 2002; Smith et al., 2002).

in incompatible elements, especially P2O5, that was found on O‘ahu shores. This erup- The lower part of Wai‘alu Ridge has the pro- Nb, Zr, Ti, and light rare earth elements. The tive activity therefore remains unconfi rmed. fi le and hummocky appearance of other sub- transition from shield to late-shield stage oc- marine rift zones (Figs. 1 and 2), although the curred as the edifi ce was beginning to emerge INTRODUCTION overall profi le is unusually shallow for its dis- from the sea. Geological observations and tance from the O‘ahu shoreline (Fig. 2), and it K/Ar ages indicate that Ka‘ena emerged A large region of shallow bathymetry appears to radiate away from the north-central above sea level ca. 3.5 Ma, reaching a maxi- extends almost 100 km WNW from Ka‘ena part of Ka‘ena Ridge, and not from either of the mum height of ~4000 m above the abyssal Point, the western tip of the island of O‘ahu, sub aerial volcanoes of O‘ahu (Fig. 1). Ka‘ena ocean fl oor and 1000 m above sea level. Rela- Hawai‘i (Fig. 1). First revealed in early bathy- Ridge has a much smoother profi le, is generally tively weak gravity anomalies, topographic metric mapping around the Hawaiian Islands shallower at a given distance from shore than lineaments, and the orientation of dike com- (e.g., U.S. Navy, 1939), it was referred to by other ridges, and is surmounted by two topo- plexes indicate a volcanic structure that is Stearns (1978) as the Waho Shelf but desig- graphic highs (Fig. 2). independent of Wai‘anae Volcano. Thus, nated Ka‘ena Ridge on later maps (e.g., Wilde Resolution of the question whether Ka‘ena volcanic structure, geochemistry, and age all et al., 1980). Most submarine ridges in the and Wai‘alu Ridges represent submarine exten- indicate a precursor volcano to the island of Hawaiian chain are extensions of volcanic rift sions of Wai‘anae Volcano, or possibly one zones (e.g., Fiske and Jackson, 1972; Smith or more previously unrecognized volcanoes, †E-mail: sinton@ hawaii .edu. et al., 2002; Lipman and Calvert, 2011), but the has important implications for understanding § Current address: U.S. Embassy, Unit 3240 Box #7, relation of Ka‘ena Ridge to the rest of O‘ahu is volcanic construction of the young end of the DPO, AA 34021-0007, Managua, Nicaragua. #Current address: Environmental Science Inter- uncertain. Ka‘ena Point (Fig. 1) is composed Hawaiian chain, and especially for the struc- national Inc., Kailua, Hawaii 96734, USA. of lava fl ows of the Wai‘anae Volcano, the tural and geochemical evolution of the island of *Deceased. older of the two volcanoes that make up the O‘ahu. Volcano spacing among the 17 youngest

GSA Bulletin; Month/Month 2014; v. 1xx; no. X/X; p. 1–26; doi: 10.1130/B30936.1; 14 fi gures; 6 tables; Data Repository item 2014194.

For permission to copy, contact [email protected] 1 © 2014 Geological Society of America Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al.

22° 10 km Kaua‘i O‘ahu Moloka‘i 21° Maui Lāna‘i Kaho‘olawe 20°

Hawai‘i 19° −3000 160° 159° 158° 157° 156° 155° D7 −4000

J2−380 Ka‘ena Slide

22° J2−305 Wai‘alu Ridge D1 −3000 00′N J2−556

D8 T−328 T−327 D2 −2000 D5 J2−251 Ka‘ena Ridge −1000 D9

X -2000 −1000 J2−306 T−325 −1000 −3000 D6 F4

F2 −4000 J2−551 Ka‘ena Pt. J2-557 D4 2500 −1000 −30−3000 21° J2−381 J2−560 30′N 21° 34′ Wai‘anae Slump

3000 3 km T−326 21° −3500 −2000 32′ 158°40′ 158°38′ 159°00′W 158°30′W Figure 1. Bathymetric map of the seafl oor west of O‘ahu, Hawai‘i; contour interval = 200 m, labeled every 1000 m. Box in upper-left inset shows map location relative to main Hawaiian Islands. Dredge tracks for R/V Kilo Moana cruise KM06–03 (prefi x D) and R/V Farnella cruise F11–88-HW (prefi x F) are shown as thick black lines; track for D8 is beneath lower part of dive J2–556. Blue (Ka‘ena and Wai‘alu Ridge) and red (secondary alkalic) dots are new samples collected by remotely operated vehicle (ROV) Jason from this study; green circles are ROV Tiburon sample locations (prefi x T) from Coombs et al. (2004), and white dots are sample locations from ROV Jason dives 305 and 306 from Greene et al. (2010). The outlines of the Ka‘ena and Wai‘anae landslide deposits are shown with light shading. Circled X shows approximate location of the “disturbance” of 22 May 1956 (from Macdonald, 1959). Inset in lower-left corner (see box on main fi gure for location; contour interval every 500 m) shows the constructional mounds of secondary alkalic pillows and lobate lavas on the lower slopes of Ka‘ena Ridge near 158°39′W, 21°34′N.

2 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

100 km A WR Figure 2. Bathymetric profi les along sub marine 22°N ridges in the Hawaiian province. (A) Map KR showing locations of profi les along the sub- PW marine Ka‘ena Ridge (KR), Wai‘alu Ridge (WR), Pa‘uwela Ridge (PW, probable exten- sion of East Moloka‘i east rift zone), Hāna Ridge (HR), Hilo Ridge (HiR), Puna Ridge HR (PR, Kīlauea east rift zone), and Ka Lae Ridge (KL, Mauna Loa SW rift zone). (B) Profi les along ridges shown in A; horizontal scale is in km from the shoreline; vertical exaggeration is 20°N HiR ~6:1. Submarine rift zones all display continu- ous increase in depth with distance from the PR shoreline, albeit at variable slopes. In contrast, Ka‘ena Ridge has a much shallower profi le and is surmounted by two topographic highs. KL

158°W 156°W 154°W

0 B Ka‘ena Wai‘alu –2000 Hāna

Depth (m) Hilo –4000 Ka Lae Pa‘uwela Puna

120 100 80 60 40 20 0 Distance (km)

volcanoes along the Hawaiian Ridge indicates Hawaiian volcanoes. Hawaiian shield volcanoes donald, 1968; Tanaka et al., 2002; Abouchami that all but one formed within 30–70 km of their are polygenetic volcanoes with quasi-indepen- et al., 2005). next oldest neighbor; at ~150 km from Kaua‘i, dent magma systems and typical life spans on The structural and geochemical characteris- Wai‘anae represents the single departure from the order of more than a million years (Mac- tics of typical Hawaiian volcanoes allow for tests this relation. A previously unrecognized vol- donald et al., 1983; DePaolo and Stolper, 1996; of whether a specifi c volcanic edifi ce should be cano at Ka‘ena Ridge in the Kaua‘i Channel, Sherrod et al., 2007; Clague and Sherrod, 2014). classifi ed as an independent volcanic system or would remove this anomalous outlier. It might Stable, well-established magma systems lead to an extension of one nearby. These tests fall into also help to explain the uniform dike orienta- long-lived central magma conduits, generally the following categories: tions in the northwest part of Wai‘anae Volcano, with gravity anomalies attributed to the pres- (1) Internal magmatic plumbing and struc- the alignment of which could refl ect buttressing ence of dense, olivine-dominated cumulates ture. Geologic evidence for calderas and rift by a preexisting edifi ce to the NW. According to within them (Strange et al., 1965; Clague, 1987; zones, and geophysical evidence for density Robinson and Eakins (2006), Wai‘anae (includ- Flinders et al., 2010, 2013). Although some anomalies indicative of long-lived magmatic ing Ka‘ena and Wai‘alu Ridges) has the fourth adjacent Hawaiian volcanoes are chemically plumbing systems can address this criterion. largest volume in the Hawaiian Islands. Should distinct (Garcia et al., 1989; Frey and Rhodes, (2) Magmatic evolutionary history. Many this volume be partitioned into two or possibly 1993; Abouchami et al., 2005), such distinc- Hawaiian volcanoes have well-documented three different volcanoes? tions are not always present, and individual evolutionary histories. For example, Wai‘anae Addressing these questions requires consid- volcanoes can erupt a range of compositions at Volcano progressed from a strongly tholeiitic eration of criteria for distinguishing individual different stages of volcano evolution (e.g., Mac- subaerial shield stage, through a chemically

Geological Society of America Bulletin, Month/Month 2014 3 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al. transitional, late-shield stage (Macdonald, Repository1). Bathymetric, magnetic, and grav- ratio determinations of Wai‘anae volcanic rocks 1968), followed by alkalic postshield vol canism ity data were collected during cruises KM06–03 (Table 4) followed procedures of van der Zan- (Sinton, 1986; Zbinden and Sinton, 1988; Presley (6 lines) and KM11–16 (7 lines). Before our der et al. (2010). Sr, Nd, and double-spike Pb et al., 1997). study, the only samples from Ka‘ena Ridge (Galer, 1999) isotope analyses were measured (3) Age. Although adjacent volcanoes can were those reported by Coombs et al. (2004). on the University of Hawai‘i VG Sector mass overlap in age by several hundred thousand Greene et al. (2010) collected samples from the spectrometer following procedures of Mahoney years, the timing of their petrological evolution channel between Ka‘ena Ridge and the island of et al. (2005), except that one of the Pb isotope can indicate whether magmatic systems in two Kaua‘i also using the ROV Jason. Dives J2–306 measurements was made in multidynamic rather locations are coupled or independent. and especially J2–305 traversed fl ow units that than static mode. Pb concentrations for the sam- In this paper, we report newly acquired geo- likely represent distal eruptions of Ka‘ena Ridge ples analyzed for isotopes were determined by logical, petrological, and geophysical data on the (Fig. 1). isotope dilution. Isotopic measurements for R/V submarine features northwest of O‘ahu shown The locations and descriptions of 99 new sam- Farnella dredge samples from the Kaua‘i Chan- in Figure 1. New isotopic data from Wai‘anae ples are reported in the supplementary material, nel (Table 4) were determined on handpicked Volcano and Kaua‘i Channel lava fl ows also are Table S1 (see footnote 1). Sixty-three volcanic glass following methods of Mahoney et al. presented. These new results address the follow- glass samples, including glassy selvages from (1991) (see supplementary material for details ing specifi c questions: pillow lavas and individual grains in hyaloclas- [footnote 1]). (1) What is the geological evolution of Ka‘ena tites, were analyzed for major-element oxide, Replicate, unspiked K-Ar (Charbit et al., and Wai‘alu Ridges? Are they entirely sub- S, and Cl concentrations (Table 1) by electron 1998) age determinations (Table 5) were made marine, or is there evidence that parts of these microprobe at the University of Hawai‘i; R/V on separated groundmass splits of fresh samples ridges were ever emergent above sea level? Farnella glass samples were analyzed by elec- after removal of phenocrysts and xenocrysts, (2) Is there evidence for rift zones on Ka‘ena tron microprobe at the U.S. Geological Survey following the methods described in Guillou and Wai‘alu Ridges, and what is their relationship at Menlo Park (see supplementary material for et al. (1998). The method allows the total 40Ar to volcanic lineaments of Wai‘anae Volcano? analytical details [footnote 1]). Reported data content of the sample to be determined with a (3) Do geophysical anomalies on Ka‘ena are averages of individual spot analyses. For precision of about ±0.2% (2σ). See supplemen- Ridge represent a continuation of O‘ahu struc- pillow rims, up to fi ve spots were measured on tary material for details (footnote 1). tures, or do they suggest independent vol- each of two to six glass chips (n = 10–27 per Residual gravity anomalies were calculated canic centers? sample); fi ve points were measured on individ- from a compilation of marine free-air anomaly (4) What is the chemical composition and ual glass fragments in hyaloclastites. data combined with land surveys (Strange et al., age of volcanic rocks from Ka‘ena and Wai‘alu Sixty-three whole-rock samples were selected 1965) on the island of O‘ahu. Variations in the Ridges, and how do they compare to those of for major- and trace-element analysis by wave- gravity fi eld resulting from fl exural deforma- Wai‘anae and Ko‘olau Volcanoes? length-dispersive, X-ray fl uorescence spec- tion of the lithosphere by island loading were (5) How many volcanoes make up the larger trometry (XRF; Table 2) at the University of removed using a 250 km boxcar high-pass fi l- O‘ahu edifi ce, and what are their individual dis- Hawai‘i. Major elements were determined on ter (Flinders et al., 2010, 2013). Additional data tributions and volumes? fused glass discs, and trace elements were mea- processing methods are given in the supplemen- (6) What does the composition of debris in sured on pressed powder pellets. The precision tary material (footnote 1). the Wai‘anae and Ka‘ena landslide deposits and accuracy of the University of Hawai‘i XRF The thicknesses of Mn-Fe–oxide rinds on indicate about the sources and timing of these system are given in Sinton et al. (2005). rock samples were measured on surfaces cut mass-wasting events? A subset of least-altered samples was selected perpendicular to sample margins; measurement (7) What are the extent, timing, and com- for additional trace-element and Sr, Nd, and Pb ranges and maxima were recorded to the nearest position of secondary volcanic activity in the isotope analyses based on regional and strati- millimeter. Kaua‘i Channel? graphic distribution and compositional vari- ability (Table 3). Trace elements were measured RESULTS FIELD INVESTIGATIONS AND at the University of Hawai‘i by inductively NEW DATA coupled plasma–mass spectrometry (ICP-MS). Morphology and Volcanic Construction Precision for ICP-MS trace-element analyses is of Ka‘ena and Wai‘alu Ridges We acquired new data and samples from better than 5% relative for all elements except four cruises to Ka‘ena Ridge: R/V Kilo Moana Co, Pb, and Th (~6%) (supplementary material, The shallow bathymetry northwest of O‘ahu cruises KM06–03 and KM06–31 in 2006, R/V Table S2 [footnote 1]). (Fig. 1) can be divided into three morphologi- Thompson cruise TN 226 in 2008, and R/V Kilo Isotopic measurements on Ka‘ena and cal regions. East of 158°32′W, bathymetric Moana cruise KM11–16 in 2011. In addition, Wai‘alu Ridge samples (Table 3) were made contours are subparallel to the O‘ahu coastline we report new data on samples from low-lying in the University of Hawai‘i isotope facility, and gradually deepen away from Ka‘ena Point lava fl ow fi elds between O‘ahu and Kaua‘i using a combination of thermal ionization mass to ~1000 m. A terrace at ~100 m was interpreted collected in 1988 during R/V Farnella cruise spectrometry (Sr and Nd isotopes) and multicol- by Coombs et al. (2004) to mark the location of F11–88-HW. Samples were collected by dredg- lector ICP-MS (Pb isotopes; see supplemen- reef complexes drowned during glacio eustatic ing during the KM06–03 and R/V Farnella tary material for details [footnote 1]). Isotopic sea-level fl uctuations in the late Pleistocene cruises; the latter three cruises (KM06–31, TN (Moore and Campbell, 1987). A striking slope 226, and KM11–16) used the remotely operated 1GSA Data Repository item 2014194, sample lo- break near ~600 m depth, ~12 km north of cations and descriptions, analytical method details, vehicle (ROV) Jason2 to collect photographs and Mn-Fe rind thicknesses, is available at http:// Ka‘ena Point (near inferred Wai‘anae-Ko‘olau and samples during seven dives (Fig. 1; supple- www .geosociety .org /pubs /ft2014 .htm or by request contact on Fig. 3), could represent the end of mentary material Table S1 in the GSA Data to editing@ geosociety .org. Ko‘olau shield building (Coombs et al., 2004) or

4 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

TABLE 1. MICROPROBE ANALYSES: WAI‘ALU AND KA‘ENA RIDGES, KA‘ENA SLIDE, AND KAUA‘I CHANNEL GLASSES S Cl

Type n SiO2 TiO2 Al2O3 FeO* MnO MgO CaO Na2OK2OP2O5 (ppm) (ppm) Sum Wai‘alu Ridge KM7-1a h 5 51.7 2.45 14.1 10.6 0.17 6.07 10.7 2.45 0.41 0.25 1150 110 99.0 KM7-1b h 4 51.6 2.39 14.1 10.6 0.16 6.36 10.6 2.42 0.41 0.25 1070 80 98.9 KM7-1c h 4 51.0 2.36 13.6 10.6 0.15 7.75 10.6 2.31 0.39 0.23 99.0 KM7-1d h 5 50.8 2.35 13.5 10.7 0.18 7.80 10.6 2.37 0.39 0.22 900 250 98.9 KM7-1e h 3 51.3 2.37 13.9 10.5 0.16 6.62 10.7 2.46 0.39 0.25 1180 90 98.8 J380-1 p 15 52.4 2.80 13.6 10.0 0.15 6.04 10.2 2.68 0.54 0.32 1090 98.8 J380-4 p 15 52.5 2.82 13.7 9.99 0.15 5.90 10.2 2.68 0.54 0.30 1130 98.9 J380-5 p 16 52.6 2.88 13.7 10.1 0.14 5.73 10.1 2.62 0.55 0.31 1150 98.8 J380-7 p 21 52.9 2.76 13.8 9.99 0.15 5.93 10.2 2.68 0.53 0.31 1090 99.3 J380-8 p 14 53.8 2.95 14.7 10.5 0.15 4.05 9.02 2.95 0.61 0.35 1360 99.2 J380-9 p 21 52.7 2.84 13.6 10.3 0.16 5.81 9.92 2.70 0.57 0.31 1210 99.0 J380-10 p 21 52.9 2.85 13.6 10.4 0.16 6.04 9.95 2.68 0.55 0.32 1250 99.6 J380-11 p 21 53.8 2.82 13.9 10.1 0.15 5.82 10.1 2.69 0.56 0.31 1090 100.4 J380-12 p 24 52.7 2.85 14.0 10.0 0.15 5.96 10.0 2.69 0.55 0.32 1080 99.3 J380-13 p 14 53.3 2.84 13.9 10.2 0.15 5.82 10.2 2.68 0.53 0.32 1110 100.0 J380-14 p 21 52.6 2.75 13.9 10.3 0.16 6.35 10.8 2.51 0.45 0.27 910 100.2 J380-17 h 21 52.5 2.67 13.8 10.2 0.16 6.39 10.8 2.47 0.44 0.27 910 99.8 J380-20 p 21 52.6 2.82 13.7 10.4 0.15 6.16 10.5 2.55 0.48 0.28 980 99.8 Ka‘ena Ridge KM4-3a p-A 21 51.0 2.37 13.4 11.3 0.18 7.59 10.6 2.33 0.40 0.23 141 116 99.4 KM9-9a h-A 5 53.5 1.84 14.4 9.83 0.15 6.23 10.1 2.59 0.51 0.23 350 120 99.5 KM9-9b h-A 6 53.5 1.70 14.5 10.0 0.15 6.75 10.2 2.63 0.43 0.18 350 100 100.2 KM9-9c h-A 5 52.1 2.10 14.2 10.5 0.17 6.37 10.7 2.43 0.51 0.22 370 70 99.3 KM9-9d h-A 5 53.5 1.66 14.5 10.0 0.16 6.76 10.3 2.60 0.43 0.17 100.1 KM9-9e h-A 5 53.7 1.76 14.7 9.99 0.15 6.23 10.2 2.73 0.44 0.17 100.0 KM9-10a h-A 5 52.5 2.27 14.1 10.4 0.16 6.22 10.3 2.59 0.49 0.27 99.4 KM9-10b h-A 5 52.5 2.24 14.1 10.6 0.16 6.14 10.3 2.57 0.49 0.26 99.2 KM9-10c h-A 5 51.5 2.37 14.1 10.7 0.15 6.50 10.5 2.55 0.49 0.27 99.2 KM9-10d h-A 5 52.5 2.24 14.1 10.4 0.18 6.16 10.3 2.61 0.49 0.25 99.2 KM9-10e h-A 5 51.3 3.13 13.5 11.5 0.17 5.78 10.3 2.56 0.64 0.36 99.3 KM9-10f h-A 5 53.1 2.30 14.1 10.4 0.14 6.24 10.4 2.49 0.51 0.27 <50 <50 99.9 KM9-10g h-A 5 51.8 3.55 14.3 11.7 0.15 5.30 8.5 2.89 0.81 0.43 99.5 KM9-10h h-A 5 53.5 2.29 14.2 10.2 0.16 6.31 10.3 2.54 0.50 0.25 100.2 KM9-10i h-A 5 53.6 2.52 13.7 10.8 0.16 5.92 10.3 2.58 0.53 0.25 100.3 KM9-10j h-A 5 53.4 2.49 13.9 10.6 0.18 6.11 10.4 2.54 0.54 0.28 80 <50 100.4 KM9-10k h-A 5 53.2 1.72 14.5 9.98 0.16 6.63 10.2 2.60 0.44 0.18 460 50 99.6 KM9-10l h-A 5 51.9 2.40 14.2 10.7 0.15 6.67 10.6 2.59 0.50 0.29 <50 130 99.9 KM9-10m h-A 5 54.8 2.00 14.2 9.60 0.14 6.80 8.90 2.91 0.50 0.23 <50 <50 100.1 KM9-10n h-A 5 52.9 2.33 14.1 10.4 0.16 6.34 10.4 2.52 0.52 0.25 80 70 100.0 KM9-10o h-A 5 53.4 2.54 13.9 10.7 0.17 6.08 10.4 2.51 0.55 0.28 <50 80 100.4 J381-1 p-S 19 45.2 2.96 12.8 12.1 0.17 6.81 13.0 3.54 1.29 0.50 1900 98.6 J381-2 p-S 18 45.5 2.78 13.2 12.1 0.18 6.22 12.8 3.56 1.24 0.50 1970 98.4 J381-3 p-S 18 45.3 2.79 13.2 12.1 0.19 6.20 12.9 3.60 1.23 0.51 1960 98.3 J381-4 p-S 18 45.7 2.80 13.3 12.1 0.18 6.07 12.8 3.66 1.23 0.51 1960 98.7 J381-6 p-S 24 46.0 2.77 13.5 12.2 0.17 5.89 12.7 3.70 1.26 0.51 2020 98.8 J381-7 p-S 20 47.0 2.67 13.9 12.5 0.18 6.78 12.8 3.43 1.09 0.44 1890 660 100.9 J381-8 p-S 24 48.2 2.41 13.4 11.6 0.16 7.09 12.1 3.09 0.81 0.33 1670 99.3 J381-9 p-A 25 52.6 2.71 14.9 11.1 0.15 6.77 10.4 2.64 0.53 0.30 340 120 102.2 J381-10 p-A 27 51.7 2.69 14.3 10.6 0.16 6.97 10.2 2.56 0.51 0.28 310 100.1 J556-2 p-A 25 50.7 2.72 15.0 11.3 0.15 6.67 10.4 2.83 0.63 0.32 1590 120 100.7 J556-3 p-A 25 51.1 2.62 14.9 11.2 0.16 6.71 10.4 2.77 0.60 0.29 1470 130 100.9 J556-12 p-B 25 50.9 3.17 14.3 12.0 0.16 5.50 10.1 3.30 1.28 0.54 <50 310 101.2 J557-3 p-S 25 46.3 2.66 13.8 12.6 0.18 7.01 12.9 3.33 1.10 0.40 1900 620 100.7 J557-5 p-B 25 50.4 2.82 14.0 11.9 0.16 7.77 11.1 2.44 0.50 0.31 840 240 101.8 Ka‘ena slump KM2-1 p-A 5 53.9 2.18 14.2 9.89 0.15 5.63 9.41 3.01 0.64 0.29 55 60 99.4 KM2-2a h-A 5 51.7 2.88 13.4 12.0 0.17 5.59 10.1 2.57 0.49 0.29 1350 110 99.4 KM2-2b h-A 5 51.7 2.72 13.8 10.6 0.17 6.10 10.3 2.71 0.52 0.30 1208 91 99.1 KM2-2c h-A 6 51.4 2.33 13.9 11.1 0.16 6.69 11.2 2.32 0.36 0.21 1256 127 99.8 KM2-2d h-A 7 51.6 2.67 13.7 11.4 0.17 6.05 10.5 2.50 0.44 0.25 177 78 99.3 KM2-2e h-A 6 51.9 2.58 13.8 11.0 0.14 6.14 10.6 2.47 0.43 0.25 1170 130 99.5 KM2-2f h-A 6 49.6 2.99 13.9 11.6 0.17 6.36 11.2 2.74 0.52 0.31 1274 109 99.5 KM2-2g h-A 4 51.8 2.74 13.7 10.6 0.17 6.19 10.1 2.69 0.53 0.31 1131 130 99.0 Kaua‘i Channel F11-2D-2 p-S 9 44.8 2.75 14.5 12.1 0.17 6.20 12.3 3.77 1.28 0.55 1860 830 98.7 F11-4D-2 p-S 9 45.0 2.70 14.6 12.6 0.17 6.66 12.7 3.74 1.05 0.46 1790 610 99.9 Note: KM samples are glasses from R/V Kilo Moana cruise KM06-03, listed by dredge and sample number; J samples are samples collected by remotely operated vehicle (ROV) Jason, listed by dive no. and sample no. Type: h—hyaloclastite or p—pillow rim. A—Ka‘ena A chemical type; B—Ka‘ena B chemical type; S—secondary lavas. All data are averages of n individual spot analyses; oxide values are in wt%; S and Cl values are in ppm. FeO* = total iron expressed as FeO. Data for R/V Farnella samples (F11-2D, 4D) were obtained using the U.S. Geological Survey electron microprobe at Menlo Park; all other data are from the University of Hawai‘i electron microprobe.

Geological Society of America Bulletin, Month/Month 2014 5 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al.

TABLE 2. WHOLE-ROCK XRF ANALYSES: WAI‘ALU AND KA‘ENA RIDGES, KA‘ENA SLIDE, AND KAUA‘I CHANNEL Location: Wai‘alu Ridge Ka‘ena Ridge Sample: J380-1 J380-4 J380-5 J380-7 J380-8 J380-9 J380-10 J380-12 J380-16 J380-18 J380-20 KM6-1 KM9-1 Group: A AAAAAAAAAABA

SiO2 50.70 51.03 50.94 50.82 50.55 50.78 50.98 50.24 49.47 49.29 50.17 48.42 52.19 TiO2 2.34 2.41 2.41 2.38 2.42 2.35 2.45 2.41 2.10 2.12 2.09 3.19 1.94 Al2O3 11.91 12.38 12.47 12.12 11.96 11.73 12.20 12.33 11.56 11.55 11.61 13.66 14.22 FeO* 11.25 11.19 11.11 11.28 11.29 11.46 11.52 11.35 11.24 11.40 11.26 12.24 10.90 MnO 0.20 0.17 0.17 0.24 0.16 0.17 0.26 0.20 0.25 0.23 0.21 0.16 0.17 MgO 12.26 11.22 11.15 11.71 11.63 12.46 11.32 12.25 13.78 14.36 14.14 9.62 8.19 CaO 8.61 9.04 9.04 8.86 8.50 8.35 8.63 8.95 9.07 8.91 8.79 9.63 10.51

Na2O 1.67 1.75 1.74 1.70 1.85 1.78 1.77 2.46 1.34 1.40 1.36 2.68 2.29 K2O 0.47 0.55 0.49 0.46 0.49 0.48 0.50 0.51 0.33 0.34 0.34 0.68 0.39 P2O5 0.27 0.27 0.27 0.27 0.28 0.29 0.29 0.29 0.22 0.22 0.20 0.53 0.27 Sum 99.67 100.02 99.81 99.84 99.13 99.86 99.92 100.97 99.36 99.84 100.18 100.80 101.08 LOI –0.02 0.40 0.36 0.31 0.25 –0.40 –0.07 0.29 0.23 0.18 –0.04 0.50 0.27

Sc 28 29 30 27 28 26 26 29 30 28 27 25 36 V 293 305 307 303 320 295 305 294 278 278 271 285 319 Cr 986 910 940 994 846 891 827 858 1161 1087 1182 584 432 Co 67 52 55 74 58 62 75 65 81 77 74 71 54 Ni 569 454 447 523 515 605 541 477 648 649 680 324 141 Zn 113 111 111 119 115 115 122 114 109 110 109 147 107 Rb88888887556115 Sr 252 264 263 261 251 246 256 267 250 247 242 493 286 Y 30303030313132322626253127 Zr 163 160 162 161 170 167 172 165 123 122 126 244 117 Nb 13 13 14 13 13 13 14 14 10 10 11 25 8 Ba 67 66 61 68 69 55 69 44 47 30 38 128 47 (continued)

TABLE 2. WHOLE-ROCK XRF ANALYSES: WAI‘ALU AND KA‘ENA RIDGES, KA‘ENA SLIDE, AND KAUA‘I CHANNEL (continued) Location: Ka‘ena Ridge Sample: KM9-2 KM9-4 KM9-8 J381-1 J381-3 J381-4 J381-6 J381-7 J381-8 J381-9 J381-10 J381-13 Group: A A A S S S S S S A A A

SiO2 51.74 53.35 53.39 44.89 44.91 45.29 45.96 45.92 46.78 48.79 49.43 50.58 TiO2 1.95 1.63 1.65 2.56 2.36 2.46 2.72 2.40 2.09 2.22 2.22 2.01 Al2O3 14.06 14.03 14.24 11.54 11.52 11.82 13.06 12.14 12.02 11.77 11.70 12.60 FeO* 10.96 10.19 10.25 12.96 13.14 13.02 12.80 12.73 12.29 11.71 11.60 11.26 MnO 0.18 0.15 0.16 0.20 0.34 0.29 0.26 0.22 0.18 0.40 0.20 0.18 MgO 8.20 9.15 8.39 12.09 12.20 11.46 7.91 10.41 11.66 13.25 13.96 12.37 CaO 10.60 9.49 9.56 11.30 11.03 11.30 12.38 11.40 10.55 8.37 8.24 8.98

Na2O 2.21 2.42 2.35 2.47 2.48 3.01 3.26 2.38 2.32 1.81 2.18 2.17 K2O 0.40 0.38 0.39 1.41 1.34 1.42 1.60 1.41 0.74 0.44 0.45 0.43 P2O5 0.27 0.23 0.24 0.45 0.44 0.44 0.49 0.41 0.30 0.25 0.24 0.22 Sum 100.57 101.02 100.62 99.87 99.76 100.51 100.45 99.42 98.95 99.01 100.22 100.79 LOI 0.20 0.39 0.37 1.27 1.50 1.39 1.70 1.86 0.02 0.23 –0.38 0.00

Sc 35 31 31 23 22 24 23 24 25 25 24 28 V 325 274 275 292 283 283 312 308 295 259 240 252 Cr 410 552 485 592 632 576 292 543 613 891 938 761 Co 65 69 55 67 88 82 61 68 66 125 76 61 Ni 139 235 210 355 376 321 141 302 412 710 704 481 Zn 108 111 110 126 129 127 130 128 119 128 121 110 Rb 5 5 6 26 23 24 27 22 20 7 7 5 Sr 295 276 277 596 585 596 657 562 477 334 314 341 Y 282524232525272522282322 Zr 119 109 109 153 160 163 179 158 105 146 138 120 Nb 8 6 7 43 41 42 47 40 26 11 11 9 Ba 61 54 52 433 412 428 464 394 267 71 65 52 (continued)

the boundary between submarine and subaerial as wide and deepens below 1450 m to the WNW 3 and 4A) extending to maximum depths of construction of the Ko‘olau Volcano (e.g., Mark at a much steeper average slope of ~7.5°. ~4600 m below sea level (mbsl). At higher and Moore, 1987). Except for a scarp on the southern side of elevations on all sides of Ka‘ena Ridge, these The region to the west of 158°32′W consists Ka‘ena Ridge that is part of the headwall of the slopes are mantled by thin fl ows of hyaloclastic of the WNW-trending Ka‘ena and NW-trending Wai‘anae slump mass-wasting deposit, the mor- debris (Fig. 4B), varying from a few centime- Wai‘alu Ridges (Fig. 1). Below 1350 m, Wai‘alu phology of Ka‘ena and Wai‘alu Ridges appears ters to a few tens of centimeters thick. These Ridge is relatively narrow, ~15 km across, with to be primarily constructional, consisting of debris fl ows formed by remobilization of sedi- hummocky morphology typical of pillow-domi- three lithological units. (1) The NW slopes of ments from upslope. (2) The lower pillows nated ridges; the crest of Wai‘alu Ridge gradu- Ka‘ena Ridge and all of the hummocky, lower are overlain by a sequence of hyaloclastites ally deepens to the northwest with a slope of part of Wai‘alu Ridge are relatively unmodi- that form steeper (~15°) slopes, up to 450 m ~4.5°. In contrast, Ka‘ena Ridge is nearly twice fi ed constructional pillow lava sequences (Figs. high. The scarp at the top of the pillows can be

6 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

TABLE 2. WHOLE-ROCK XRF ANALYSES: WAI‘ALU AND KA‘ENA RIDGES, KA‘ENA SLIDE, AND KAUA‘I CHANNEL (continued) Location: Ka‘ena Ridge Sample: J551-2 J551-3 J551-4 J551-8 J551-11 J556-1 J556-2 J556-3 J556-4 J556-5 J556-6 J556-7 J556-8 Group: BBBBAAAAAAAAA

SiO2 46.35 46.82 46.05 48.74 50.50 49.07 50.02 50.46 50.37 53.11 52.37 48.35 51.44 TiO2 2.78 2.94 2.79 3.01 3.79 2.76 2.70 2.53 2.59 1.51 1.55 1.92 1.68 Al2O3 10.55 10.78 10.59 17.48 14.19 15.19 15.03 14.66 15.08 13.56 13.41 12.49 13.69 FeO* 12.92 12.70 13.02 11.47 13.53 11.85 11.35 11.02 11.28 10.46 10.75 12.67 11.20 MnO 0.33 0.23 1.16 0.21 0.21 0.23 0.18 0.17 0.17 0.18 0.24 1.39 0.21 MgO 13.95 13.82 13.60 3.92 4.84 6.62 6.45 6.91 6.25 9.36 9.41 10.96 9.31 CaO 9.16 8.88 9.08 7.39 8.96 10.40 10.24 10.10 10.35 8.84 9.15 9.12 9.71

Na2O 1.84 1.90 1.79 4.35 3.06 2.36 2.31 2.21 2.19 2.09 1.93 1.55 1.82 K2O 0.47 0.58 0.57 1.69 0.44 0.56 0.61 0.58 0.60 0.29 0.28 0.46 0.29 P2O5 0.47 0.50 0.43 1.16 0.50 0.33 0.31 0.28 0.32 0.14 0.16 0.24 0.16 Sum 98.83 99.16 99.08 99.43 100.02 99.37 99.22 98.93 99.20 99.55 99.26 99.14 99.51 LOI 0.63 0.23 1.33 1.16 0.20 1.40 1.30 0.61 1.11 0.50 0.26 1.89 0.56

Sc 27 26 28 12 30 29 28 30 28 31 31 27 32 V 292 283 299 137 391 328 330 317 327 267 280 325 297 Cr 1048 935 1050 17 17 355 317 348 341 671 701 709 668 Co 107 82 127 34 46 71 53 111 49 57 73 335 79 Ni 633 639 659 25 57 162 146 152 141 290 297 605 252 Zn 160 144 170 180 156 154 132 130 138 112 109 158 106 Rb79924578894474 Sr 501 513 490 1244 358 454 434 413 417 198 212 333 221 Y 32323652473128272723243526 Zr 187 202 192 483 255 174 173 160 160 91 95 136 95 Nb2323226422151414136696 Ba 144 162 168 463 75 68 85 87 86 46 41 197 26 (continued)

TABLE 2. WHOLE-ROCK XRF ANALYSES: WAI‘ALU AND KA‘ENA RIDGES, KA‘ENA SLIDE, AND KAUA‘I CHANNEL (continued) Location: Ka‘ena Ridge Sample: J556-9 J556-10 J556-11 J556-17 J557-2 J557-3 J557-4 J557-5 J557-7 J557-9 J557-11 J560-1A J560-1B Group: A A A B S S S B B B A A A

SiO2 51.01 50.23 50.36 47.21 45.90 45.20 45.34 47.64 47.78 46.25 50.32 50.48 50.46 TiO2 1.54 1.62 1.55 2.81 2.39 2.34 2.46 2.47 2.47 2.46 2.68 1.93 1.93 Al2O3 13.76 14.34 13.45 14.61 12.16 12.05 11.99 12.15 12.34 11.47 13.87 13.24 13.19 FeO* 11.16 11.34 11.23 12.97 13.03 12.85 13.38 12.47 12.25 13.37 10.78 11.02 10.98 MnO 0.50 0.20 0.44 0.29 0.21 0.29 0.29 0.16 0.17 0.23 0.16 0.18 0.17 MgO 9.83 9.40 9.88 6.47 12.51 11.64 12.27 12.37 11.88 14.06 7.66 10.05 10.01 CaO 9.47 10.05 8.98 10.36 10.95 11.13 11.24 9.62 9.65 9.16 10.88 9.96 9.92

Na2O 1.99 2.04 2.73 2.69 2.44 2.54 2.39 1.79 1.75 1.72 1.95 1.92 1.89 K2O 0.31 0.24 0.36 1.07 1.13 1.23 1.04 0.43 0.46 0.37 0.57 0.27 0.28 P2O5 0.17 0.17 0.19 0.76 0.38 0.41 0.39 0.25 0.30 0.29 0.33 0.24 0.23 Sum 99.74 99.62 99.18 99.24 101.10 99.69 100.79 99.36 99.05 99.37 99.21 99.30 99.07 LOI 1.00 0.37 0.80 2.02 0.98 1.41 1.13 0.95 0.05 1.10 0.66 0.67 0.97

Sc 31 33 30 29 20 21 20 28 33 25 32 31 33 V 289 313 297 432 302 312 311 282 292 274 301 291 293 Cr 628 642 756 730 615 559 585 800 784 892 428 718 675 Co 170 57 124 78 73 81 88 67 115 87 45 77 63 Ni 335 259 337 265 409 347 404 532 469 641 192 255 253 Zn 114 130 136 146 121 126 127 131 135 145 110 121 112 Rb54512242621887944 Sr 247 248 233 646 573 521 572 311 305 343 374 258 255 Y 28272731232324282729312423 Zr 93 85 101 194 126 121 129 150 149 148 174 107 109 Nb 6 6 7 27 36 34 35 16 16 16 17 9 8 Ba 54 35 63 299 442 437 442 66 90 68 95 48 39 (continued)

followed in the bathymetry between the loca- sisting of gravel and cobble beach deposits and J2–557, while pillows, highly vesicular lava, tion for dive J2–556 and Wai‘alu Ridge (Fig. 1), massive ‘a‘ā lava fl ows. and hyaloclastite were all recovered in KM06 although it shoals and thins to the east. (3) The The transition from submarine to subaerial dredge 9 between 2400 and 1600 m depth. The gently sloping (3°–5°) upper reaches of Ka‘ena volcanism is constrained by observations at sev- vesicular lava fragments have abundant spheri- Ridge are dominated by fi ne-grained, locally eral locations. The boundary between pillows cal vesicles and lack glassy selvages, which remobilized, epiclastic sediments in layers a and overlying hyaloclastites coincides with we interpret to indicate subaerially emplaced few decimeters thick, with rare exposures of a sharp break in slope at ~1850 mbsl on dive pāhoehoe lava. Dive J2–551, which began at volcanic rubble and subaerially erupted lava J2–556 (Fig. 1). A similar slope break character- a depth of 2100 m, mainly traversed hyalo- (Fig. 4C). Coombs et al. (2004) described the izes the upper parts of Wai‘alu Ridge, 20 km to clastites and a sequence of massive lava fl ows summit of the eastern bathymetric high visited the north. On the south side of Ka‘ena Ridge, lacking pillow structure. Recovered samples are during ROV Tiburon dive T-325 (Fig. 1) as con- pillows were recovered below 3000 m on dive devoid of marginal glass or radial fractures and

Geological Society of America Bulletin, Month/Month 2014 7 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al.

TABLE 2. WHOLE-ROCK XRF ANALYSES: WAI‘ALU AND KA‘ENA RIDGES, KA‘ENA SLIDE, AND KAUA‘I CHANNEL (continued) Location: Ka‘ena Ridge Ka‘ena slide Kaua‘i Channel Sample: J560-5 J560-6 J560-7 J560-8 J560-9 J560-10 J560-11 J251-1 J251-2 J251-3 F2-2 F4-2 Group: AAAAAABAAASS

SiO2 47.87 48.76 50.82 49.56 45.51 49.32 45.56 51.41 50.92 51.86 45.67 44.91 TiO2 2.23 2.25 2.30 2.18 1.24 2.13 2.66 2.55 2.81 2.64 2.83 2.37 Al2O3 12.55 12.67 13.61 12.74 7.36 14.00 14.11 14.12 13.43 14.41 14.04 12.35 FeO* 11.33 11.36 11.00 10.93 11.90 11.23 13.59 11.29 11.51 11.37 12.26 13.04 MnO 0.16 0.16 0.16 0.16 0.15 0.16 0.22 0.16 0.17 0.18 0.18 0.19 MgO 13.07 12.69 8.57 11.79 28.01 10.00 10.19 7.05 9.09 6.70 6.52 11.94 CaO 8.51 8.79 10.46 9.77 4.49 9.80 10.37 10.79 9.42 10.96 13.21 11.43

Na2O 2.76 2.05 1.98 1.84 1.05 2.85 2.14 2.14 2.31 2.24 3.06 3.12 K2O 0.46 0.44 0.38 0.28 0.16 0.30 0.24 0.53 0.56 0.51 2.07 0.96 P2O5 0.24 0.28 0.25 0.26 0.18 0.22 0.38 0.25 0.31 0.25 0.55 0.41 Sum 99.20 99.46 99.69 99.51 100.05 100.02 99.46 100.30 100.53 101.12 100.39 100.72 LOI 0.67 0.39 0.66 –0.16 –0.19 0.48 0.75 0.66 0.82 0.55 2.57 0.61

Sc 22 28 32 33 16 30 32 33 30 30 21 22 V 245 269 310 276 138 278 349 335 345 336 331 299 Cr 753 751 554 815 1261 578 697 345 618 277 97 433 Co 72 64 60 56 156 57 96 43 49 41 45 66 Ni 597 560 201 417 1779 277 447 95 302 72 90 364 Zn 114 115 110 107 116 120 144 112 127 115 122 119 Rb67742429983925 Sr 348 331 304 305 203 334 364 293 293 300 699 534 Y 242426251325302936302620 Zr 141 141 135 129 81 127 170 153 194 155 150 125 Nb 12 12 12 11 6 10 16 14 17 14 43 34 Ba 56 85 51 65 42 26 77 52 75 52 609 380 Note: All analyses were conducted by X-ray fluorescence (XRF) at the University of Hawai‘i; major-element oxides are in wt%; trace elements are in ppm. FeO* = total Fe expressed as FeO. LOI—loss on ignition at 900°C. Group designations: A—Ka‘ena A; B—Ka‘ena B; S—secondary (discussed in the text).

TABLE 3. ICP-MS AND ISOTOPIC DATA: WAI‘ALU AND KA‘ENA RIDGES Location: Wai‘alu Ridge Ka‘ena Ridge Sample: J380-1r J380-8r J380-18r J381-1r KM6-1r J381-6r J381-8r J381-10r J551-2r Group: A A A S B S S A B Li 6.8 7.4 5.6 5.2 8.8 5.4 5.6 6.1 7.6 Co 42 39 43 60 43 47 53 49 60 Ni 188 118 235 296 186 87 195 223 364 Rb 8.3 9.2 6.1 28.8 9.7 28.8 21.1 7.6 10.5 Sr 295 307 286 641 503 715 537 386 528 Y 343829243129242931 Zr 183 202 148 157 253 188 111 164 214 Nb 15.0 16.4 12.6 45.3 24.9 53.8 25.4 13.1 25.2 Cs 0.08 0.10 0.07 0.39 0.10 0.38 0.27 0.08 0.12 Ba 89 96 70 489 137 507 324 96 174 La 13.0 14.2 10.8 29.1 21.1 33.3 18.2 11.7 21.3 Ce 33 37 28 57 48 65 39 30 49 Pr 4.6 5.1 3.9 7.0 6.9 7.9 4.8 4.2 6.8 Nd 22.6 24.6 19.4 29.7 32.2 33.6 21.4 20.5 31.5 Sm 6.3 6.9 5.4 6.7 8.0 7.6 5.4 5.6 7.8 Eu 2.05 2.22 1.79 2.26 2.65 2.52 1.90 1.90 2.52 Gd 6.6 7.1 5.7 6.5 8.0 7.4 5.5 5.8 7.7 Tb 1.06 1.18 0.92 0.97 1.19 1.12 0.87 0.94 1.18 Dy 6.3 7.1 5.5 5.0 6.4 5.9 4.8 5.4 6.2 Ho 1.19 1.34 1.05 0.87 1.14 1.03 0.86 1.02 1.10 Er 3.16 3.53 2.82 2.14 2.87 2.54 2.13 2.67 2.70 Tm 0.44 0.47 0.38 0.27 0.38 0.33 0.28 0.36 0.35 Yb 2.61 2.90 2.30 1.63 2.28 1.94 1.68 2.17 2.06 Lu 0.37 0.42 0.33 0.22 0.31 0.27 0.23 0.31 0.28 Hf 4.7 5.2 3.9 3.9 6.4 4.5 3.0 4.2 5.4 Pb 1.1 1.3 0.9 2.4 1.7 2.3 1.4 1.2 1.3 Th 0.94 1.00 0.77 3.40 1.80 3.93 2.00 0.77 1.72 U 0.31 0.33 0.24 0.89 0.51 1.02 0.56 0.25 0.51

206Pb/204Pb 18.179 18.173 18.283 18.610 18.345 18.225 18.056 18.633 207Pb/204Pb 15.457 15.456 15.462 15.503 15.469 15.460 15.453 15.512 208Pb/204Pb 37.927 37.923 37.954 38.163 38.051 37.846 37.849 38.200 143Nd/144Nd 0.51295 0.51294 0.51301 0.51295 0.51299 0.51303 0.51292 0.51304 87Sr/86Sr 0.703651 0.703702 0.703378 0.703651 0.703285 0.70324 0.703666 0.703514

εNd 6.0 5.9 7.2 6.0 6.9 7.6 5.4 7.8 Pb-ID (ppm) 1.16 0.92 2.15 1.33 2.29 1.27 1.10 1.44 (continued)

8 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

TABLE 3. ICP-MS AND ISOTOPIC DATA: WAI‘ALU AND KA‘ENA RIDGES (continued) Location: Ka‘ena Ridge Sample: J551-3r J551-8r J551-11r J556-3g J556-9r J556-17r J557-3g J560-7r J560-10r Group: B B A A A B S A A Li 7.9 15.4 12.3 6.5 7.7 8.9 5.6 13.0 11.1 Co 64 23 43 47 37 34 67 40 43 Ni 383 10 55 144 125 139 411 120 166 Rb 12.1 30.4 5.6 8.7 4.9 13.1 26.4 5.3 3.5 Sr 605 1256 364 421 240 719 531 312 344 Y 355047262330222827 Zr 245 498 262 166 107 207 129 141 141 Nb 24.2 66.4 21.3 12.5 6.1 30.3 33.9 11.8 9.7 Cs 0.15 0.29 0.06 0.09 0.08 0.02 0.33 0.19 0.17 Ba 199 466 103 103 49 319 396 53 56 La 22.9 49.4 18.4 12.1 7.4 28.6 22.4 11.9 10.4 Ce 53 116 45.7 30 17 61 44 26 25 Pr 7.2 16.0 6.6 4.4 2.5 8.2 5.6 4.1 3.7 Nd 33.6 70.0 32.3 20.7 12.3 35.3 23.5 19.5 18.2 Sm 8.7 15.7 9.1 5.6 3.5 8.0 5.5 5.3 5.1 Eu 2.79 4.97 2.96 1.91 1.29 2.65 1.91 1.84 1.77 Gd 8.3 14.5 9.4 6.0 4.4 8.5 5.7 6.0 5.5 Tb 1.28 2.13 1.54 0.89 0.65 1.15 0.84 0.89 0.84 Dy 6.9 10.6 8.9 5.1 4.1 6.1 4.5 5.2 5.0 Ho 1.21 1.83 1.69 0.94 0.83 1.06 0.80 1.00 0.96 Er 3.00 4.46 4.30 2.46 2.24 2.70 2.00 2.62 2.53 Tm 0.39 0.57 0.59 0.33 0.32 0.36 0.26 0.36 0.36 Yb 2.29 3.40 3.59 1.99 1.96 2.14 1.57 2.20 2.13 Lu 0.32 0.47 0.52 0.28 0.28 0.29 0.21 0.31 0.30 Hf 6.2 11.1 6.9 4.2 2.9 5.2 3.5 3.8 4.1 Pb 1.7 3.3 1.5 2.3 1.1 2.9 1.6 1.1 1.1 Th 1.79 4.24 1.41 0.90 0.66 2.26 2.73 0.82 0.79 U 0.63 1.36 0.31 0.25 0.14 1.23 0.68 0.40 0.17

206Pb/204Pb 18.610 18.581 18.373 17.928 17.870 18.231 18.048 18.155 17.929 207Pb/204Pb 15.506 15.494 15.493 15.451 15.448 15.466 15.476 15.466 15.454 208Pb/204Pb 38.171 38.132 38.063 37.777 37.754 37.877 37.843 37.929 37.838 143Nd/144Nd 0.51303 0.51303 0.51296 0.51286 0.51284 0.51303 0.51289 0.51294 0.51288 87Sr/86Sr 0.703539 0.703478 0.703867 0.703827 0.703934 0.703257 0.703916 0.703755 0.703782

εNd 7.6 7.7 6.3 4.4 3.9 7.5 4.8 5.9 4.7 Pb-ID (ppm) 1.40 3.19 1.46 1.32 0.88 1.54 2.67 0.98 1.04 Note: Sample designations: r—rock, g—glass. Group designations are same as in Table 2. ICP-MS—inductively coupled plasma–mass spectrometry. Pb-ID are Pb abundances (ppm) determined by isotope dilution. Trace elements are in ppm.

have vesicles that range from spherical to elon- depth. Multi ple dikes, ~0.5 m wide and striking wall scarp and the presence of tilted fault blocks gate; we interpret these samples to represent ~N40W ± 20°, were encountered up to depths within it. However, Moore et al. (1989) noted subaerially erupted lava fl ows. Similar massive as shallow as ~1850 m in this region. The dikes that it shows some characteristics of a slump, lava fl ows lacking glassy selvages occur above intrude breccias that represent either rubbly fl ow being wide relative to its length, and it meets 1650 m on dive J2–560. The combined observa- units or talus deposits (Fig. 4D). The J2–560 the deep seafl oor at a steep angle. Robinson and tions suggest that Ka‘ena Ridge evolved from dike complex lies ~10 km south of the extension Eakins (2006) treated this deposit as a slump submarine to subaerial volcanism and that the of the N60W-trending Wai‘anae NW rift zone associated with Wai‘anae Volcano. In this paper, present thickness of subaerially erupted lava (Fig. 3). we refer to this feature as the Ka‘ena slide. on Ka‘ena Ridge is slightly less than 1000 m. Dikes also occur at ~1760 m depth on dive Two shield-shaped topographic highs that sur- J2–551, but these strike NNE, suggesting they Secondary Volcanism mount the ridge do not appear to be signifi cantly are associated with the shallow topographic eroded, suggesting the present thickness of sub- shield on Ka‘ena Ridge sampled by Tiburon dive Lava fl ow fi elds with highly refl ective back- aerial lava approximates the maximum eleva- T-325; this fi nding is consistent with the inter- scatter characteristics were discovered in the tion above sea level attained at Ka‘ena Ridge pretation of Smith (2002) that the SSW-trending channel between O‘ahu and Kaua‘i during prior to subsidence. ridge could represent a subsidiary rift zone. GLORIA side-scan sonar surveys (Torresan et al., 1989; Holcomb and Robinson, 2004). Sev- Dike Complexes and Rift Zones Landslide Deposits eral of these features were subsequently sampled during R/V Farnella cruise F11–88-HW (Clague Rift zones comprise the locus of volcanic Two fi elds of landslide debris are present in et al., 1988), ROV Kaiko and Shinkai 6500 sub- eruptions and feeder dikes; the delineation of the area (Figs. 1 and 3). The apparent presence mersible dives K204 and S706 (Nakagawa and rift zones in eroded volcanoes is based on ori- of multiple headwall scarps for the northern part Noguchi, 2003; Hanyu et al., 2005), and by the entation and abundance of dikes (Walker, 1986; of the Wai‘anae slump suggests that this deposit ROV Jason (Greene et al., 2010). The 40Ar-39Ar Zbinden and Sinton, 1988) in addition to the formed during multiple slide events (Moore ages for several different lava fl ows cluster in the distribution of eruptive vents. We observed et al., 1989). The deposit north of Ka‘ena Ridge range from ca. 0.4 Ma (Nakagawa and Noguchi, dikes in two locations on Ka‘ena Ridge. Dive was called the Ka‘ena debris avalanche by 2003) to 0.37 Ma (Greene et al., 2010), with J2–560 started in a dike complex at 2500 m Moore et al. (1989) on the basis of a single head- one sample yielding an age of 1.37 Ma (Greene

Geological Society of America Bulletin, Month/Month 2014 9 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al.

TABLE 4. ISOTOPIC DATA FOR WAI‘ANAE VOLCANO AND KAUA‘I CHANNEL LAVA FLOWS 5 and 6) and low FeO* and incompatible ele- 206 204 207 204 208 204 143 144 87 86 Sample Pb/ Pb Pb/ Pb Pb/ Pb Nd/ Nd Sr/ Sr εNd Pb-ID ments. Chondrite-normalized (La/Sm)N is 1.3– Lualualei 1.5, and (Sm/Yb) is 2.0–3.5 (Fig. 7). Ka‘ena OW-161 18.100 15.459 37.890 0.512968 0.703612 6.4 1.16 N OW-175 18.173 15.462 37.867 0.512970 0.703592 6.4 1.04 B samples range from tholeiitic to alkalic basalt OW-183 18.016 15.452 37.747 0.512939 0.703728 5.8 1.64 and hawaiite (one sample) (Fig. 5). Compared OW-185 18.001 15.447 37.822 0.512939 0.703696 5.8 0.77 to Ka‘ena A samples, for a given MgO, Ka‘ena Kamaile‘unu B samples have lower SiO , and higher FeO* OW-97 18.154 15.471 37.849 0.512959 0.703650 6.2 1.26 2 OW-99 18.219 15.483 37.904 0.512975 0.703580 6.5 1.36 and many incompatible elements, most notably OW-100 18.108 0.512969 0.703523 6.4 1.58 P, Nb (Fig. 6), Zr, and Ti (not shown), and range OW-198 18.110 15.466 37.790 0.512945 0.703696 5.9 1.77 OW-206 18.122 15.460 37.815 0.512944 0.703631 5.9 1.68 to higher CaO/Al2O3 and Ba/V (Fig. 6). Com- OW-217 18.053 15.453 37.803 0.512979 0.703605 6.6 1.91 pared to Ka‘ena A, Ka‘ena B are enriched in

Pālehua light rare earth elements (REEs) with (La/Sm)N

OW-203 18.162 15.469 37.836 0.512939 0.703710 5.8 3.66 ratios in the range 1.5–2.5, and (Sm/Yb)N = 3.5– C-170 18.048 15.460 37.759 0.512932 0.703736 5.7 2.44 C-177 18.070 15.463 37.777 0.512926 0.703735 5.6 2.65 5.5 (Fig. 7). Although only high-MgO samples C-182 18.133 15.462 37.828 0.512948 0.703686 6.0 3.45 were recovered from Wai‘alu Ridge, those sam- Kolekole ples are close to Ka‘ena A in all compositional OW-77 18.172 15.476 37.856 0.512939 0.703739 5.8 2.16 characteristics for their MgO contents (Figs. OW-147 18.174 15.470 37.831 0.512949 0.703764 6.0 1.64 5–7). Secondary lavas range from transitional OW-230 18.144 15.466 37.809 0.512940 0.703721 5.9 1.80 PW-29 18.156 15.471 37.834 0.512931 0.703787 5.7 1.72 basalt to basanite, are enriched in Nb and most PW-89 18.078 15.461 37.805 0.512936 0.703697 5.8 1.50 other incompatible elements, have the highest Kaua‘i Channel CaO/Al2O3, and are the most depleted in heavy F11, 2-2 18.129 15.433 37.723 0.513042 0.703304 7.8 1.95 REEs (Figs. 5–7). F11, 4-2 18.096 15.432 37.694 0.513088 0.703150 8.7 1.51 Note: Wai‘anae samples are grouped by stratigraphic member of the Wai‘anae Volcanics (Sinton, 1986; All pillow samples from the lower slopes of Presley et al., 1997). Pb-ID—Pb abundances (ppm) determined by isotope dilution. Wai‘anae data are on Ka‘ena and Wai‘alu Ridges, including those whole-rock samples; see van der Zander et al. (2010) for procedures, standards, blanks, and estimated errors; from dives J2–305 and J2–306 (Greene et al., Pb isotopic ratio measurements utilized the double-spike method (Galer, 1999). F11 dredge samples are from the R/V Farnella cruise F11-88-HW. These samples were run at the same time as those reported by Frey et al. 2010) and those from KM dredge 9, have (2000) using procedures, standards, fractionation corrections, and estimated errors reported therein. See Ka‘ena A compositions, whereas group B com- supplementary material for additional information (text footnote 1). prises mainly subaerial lavas. The distribution of Ka‘ena A and Ka‘ena B samples is not strictly stratigraphic, however. For example, although et al., 2010). Thick palagonite rinds on some of do on the channel fl oor. Rather they are similar dive J2–551 is dominated by Ka‘ena B compo- the Farnella samples are consistent with ages to peripheral secondary alkalic volcanism such sitions, the shallowest sample collected during older than 1 Ma (Moore et al., 1985). as that found at the South Arch and North Arch that dive (J551–11) is a differentiated variant We discovered a previously unknown fi eld of volcanic fi elds in the Hawaiian province (Lip- (4.84 wt% MgO) of Ka‘ena A magma. It has the young volcanic rocks mantling the south side man et al., 1989; Clague et al., 1990). highest total REE abundances but a REE pattern of Ka‘ena Ridge (Fig. 1, inset) during dives that is parallel to most other Ka‘ena A samples J2–381 and J2–557. This fi eld appears to have Chemical and Isotopic Compositions (Fig. 7). Samples collected by ROV Tibu- been erupted from two constructional mounds, ron from the shallowest part of Ka‘ena Ridge each several hundred meters high. The deeper We collected ~100 new samples from Ka‘ena (T-325) include both Ka‘ena A and Ka‘ena B mound erupted basanite pillow lava (Tables 1 and Wai‘alu Ridges, the Ka‘ena slide, and the compositions. Taken together, the sample data and 2; J381 samples 1–4), whereas the shal- apparently much younger secondary lavas that suggest that the evolution of Ka‘ena Ridge lower, apparently younger mound erupted drape the southernmost slopes of Ka‘ena Ridge involved an early period when only Ka‘ena A mildly alkalic basalt (sample J381–8) with and extend out into the channel between O‘ahu compositions were erupted, followed by a later lobate lava morphology. Figure 3 shows the and Kaua‘i (Fig. 3). Ka‘ena Ridge samples show phase characterized by eruption of both Ka‘ena distribution of presently known, young, alkalic a range of compositions, which we have divided A and Ka‘ena B chemical types. volcanism between O‘ahu and Kaua‘i. These into two main groups. Ka‘ena A samples are Glass sulfur contents generally correlate with volcanic fi elds do not appear to represent reju- mostly tholeiitic (one transitional basalt, Fig. 5), total alkalis (Fig. 8) and FeO* (not shown), venation of a preexisting volcano, lying as they with relatively high SiO2 (48–55 wt%; Figs. but only for S contents >800 ppm. All glass

TABLE 5. AGE DETERMINATIONS, KA‘ENA RIDGE K Mass molten 40Ar* Weighted mean 40Ar* Age Sample Group (wt%, ±1σ) (g) 40Ar* (%) (×10–12 mol/g, ±1σ) (×10–12 mol/g, ±1σ) (Ma, ±2σ) KM 6-1 KR-B 0.573 ± 0.006 1.02527 7.705 3.569 ± 0.019 3.560 ± 0.014 3.58 ± 0.08 1.00115 8.493 3.549 ± 0.020 KM 9-1 KR-A 0.332 ± 0.003 1.10747 13.100 2.107 ± 0.012 2.074 ± 0.009 3.60 ± 0.08 1.14362 6.472 2.050 ± 0.012 KM 9-4 KR-A 0.340 ± 0.003 1.02954 11.663 1.474 ± 0.009 1.470 ± 0.007 2.49 ± 0.06 1.46978 5.385 1.464 ± 0.011 J551-11 KR-A 0.423 ± 0.004 1.19182 8.235 2.066 ± 0.024 2.096 ± 0.014 2.86 ± 0.05 0.91775 7.602 2.113 ± 0.018 Note: All samples are from Ka‘ena Ridge; precise locations are given in supplementary material Table S1 (see text footnote 1); KR-B—Ka‘ena group B; KR-A—Ka‘ena group A.

10 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

-4000 Legend

N secondary lava

landslides

Ka‘ena and Wai‘alu Ridges -3500 -4500 22° subaerial lava 00′N -2500 4000 hyaloclastite and Ka‘ena Slide *# epiclastic sediment

-3000 pillow lava

-1000 -1500 gravity anomalies -3000 -2000

*#*# rift zones ? -1000 -3000 Ko‘olau -500 -4000 scarps Wai‘anae

O‘ahu 01020 21° -4500 500 30′N Kilometers Wai‘anae Slump -3500

-2500 *#-2000

159°10′W 158°50′W 158°30′W 158°10′W

Figure 3. Geologic map of Ka‘ena and Wai‘alu Ridges, based on observations and samples of this study combined with analysis of seafl oor texture derived from shaded relief images and backscatter characteristics. Much of the region shown as subaerial lava is covered by redis- tributed epiclastic sediment. Possible rift zones are shown as bold dashed lines. The outlines of two gravity anomalies from Figure 10 are shown as coarse dotted lines. The inferred location of the Wai‘anae-Ko‘olau contact in the offshore region is shown by a thin dashed line. See text for discussion.

samples from Wai‘alu Ridge recovered dur- These data indicate that the transition from and higher values of FeO, P2O5, Nb, and Ba/V ing dive J2–380 and KM06 dredge 7 at depths Ka‘ena A to Ka‘ena B occurred after Ka‘ena of Ka‘ena B, compared to Ka‘ena A, are broadly >2290 m and some samples from the Ka‘ena Ridge reached shallow water depths or pos- similar to differences between postshield and slide have ≥900 ppm S (Table 1; Fig. 8), indi- sibly after emergence above sea level. All but shield data from the Wai‘anae Volcano (Figs. 5 cating compositions that retained dissolved gas one secondary lava samples have >1670 ppm and 6). However, Ka‘ena B tends to have simi- from deep-water eruptions (Dixon et al., 1991). S (J557–3 has 620 ppm S), consistent with the lar or higher CaO/Al2O3 compared to Ka‘ena A, Ka‘ena Ridge samples J556–2 and J556–3, col- restriction of these lava fi elds to depths greater whereas Wai‘anae postshield lavas tend to have lected between 2628 and 2698 m depth, have than 2340 m (Fig. 3). lower CaO/Al2O3 ratios than Wai‘anae shield Ka‘ena A compositions and high sulfur con- There is considerable overlap in chemical samples (Fig. 6E). tents of 1470 and 1590 ppm S, respectively. composition among the Wai‘anae and Ko‘olau Ka‘ena and Wai‘alu Ridge sample data defi ne Other Ka‘ena A glasses have lower sulfur con- shields and our sample data from Ka‘ena and relatively tight correlations for most radiogenic tents that range from 30 to ~500 ppm, indicat- Wai‘alu Ridges. Ka‘ena A and Wai‘alu samples isotope ratios over a large range in 206Pb/204Pb, ing that some Ka‘ena A eruptions occurred are essentially indistinguishable from each other from 17.87 to 18.63 (Table 3; Fig. 9), one of subaerially or in shallow water. These include and plot on the high-SiO2, low-FeO*, and low- the largest ranges in Pb isotope composition in samples in KM06 dredges 4 and 9, J381–9 and P2O5 side of the data fi elds for Wai‘anae shield the Hawaiian Islands (e.g., Weis et al., 2011). 206 204 J381–10 (loose samples, not from outcrops), samples, but within the fi elds for Ko‘olau sam- Low- Pb/ Pb, relatively high-SiO2 Ka‘ena A and some samples from the Ka‘ena slide. Glass ples on Figures 5 and 6. Compared to Wai‘anae samples have compositions that overlap those was recovered from only two Ka‘ena B samples and Ko‘olau data at the same MgO, Ka‘ena from Ko‘olau Volcano, whereas the very high- 206 204 (J556–12, 1946 m, and J557–5, 3228 m, talus A and Wai‘alu samples tend to have slightly Pb/ Pb and high-εNd Ka‘ena B samples sample likely derived from higher up), and both higher Sc and V, and lower Ba values, with con- have values comparable to those from Kīlauea, have low S (36 and 237 ppm S, respectively). sequently lower Ba/V (Fig. 6F). Lower SiO2 Mauna Kea, and Hāna Ridge (Ren et al., 2006).

Geological Society of America Bulletin, Month/Month 2014 11 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al.

A B

C D

Figure 4. Photographs showing geologic units described in the text. (A) Pillow lavas, dive J2–556, 2110 m depth. (B) Hyaloclastite sheets, ~10 cm thick, dive J2–557, 2878 m depth. (C) Subaerial lava at top of slope dominated by hyaloclastite, dive J2–556, 1363 m depth. (D) Dike cutting breccia, dive J2–560, 2068 m depth. Width of each photograph is ~1.5–2.5 m.

The well-correlated Ka‘ena arrays suggest that A subset of Wai‘anae samples has isotopic KM06–03 (6 lines) and KM11–16 (7 lines). much of the variation can be explained by mix- ratios that are displaced to higher 208Pb/204Pb These data represent the fi rst comprehensive ing between two primary source components. and lower εNd compared to the main Wai‘anae geophysical survey of the region immedi- In contrast to the well-correlated variation fi eld (Ws, Fig. 9). These include silicic lavas ately west of O‘ahu. A map of residual grav- 206 204 over an extreme range of Pb/ Pb and εNd val- and dikes (Sinton, 1986; Zbinden and Sinton, ity anomalies compiled using these and data 206 204 ues exhibited by the Ka‘ena and Wai‘alu sam- 1988) with the lowest Pb/ Pb and εNd of any from other cruises in the region (Flinders et al., ples, the Wai‘anae isotopic data show more scat- measured samples in Wai‘anae Volcano (van der 2013), following corrections for topography and ter over a smaller range of isotopic ratios (Fig. 9; Zander et al., 2010). fl exural deformation similar to Flinders et al. Table 4). Ka‘ena and Wai‘alu data form a trend Secondary lava samples have isotopic com- (2010), is shown in Figure 10. The gravity fi eld 206 204 with lower εNd for a given Pb/ Pb than most positions (Fig. 9) similar to rejuvenated-stage is dominated by the large-amplitude Wai‘anae Wai’anae data, and this trend is highly oblique volcanics from Kaua‘i (Kōloa), O‘ahu (Hono- anomaly, which coincides with the geologically to that of Ko‘olau Volcano (Fig. 9C). Most lulu), West Maui (Lahaina), and lava fl ows from defi ned caldera region (Sinton, 1986; Zbinden Wai‘anae samples also have lower 208Pb/204Pb the North Arch volcanic fi eld, particularly in and Sinton, 1988; van der Zander et al., 2010). 206 204 87 86 values for a given Pb/ Pb than the Ka‘ena- having low Sr/ Sr and high εNd at intermediate The Wai‘anae anomaly is elongate to the north- Wai‘alu trend (Fig. 9B). In terms of 208Pb/204Pb 206Pb/204Pb (Stille et al., 1986; Frey et al., 2000; west, crossing the shoreline 6–7 km southeast versus 206Pb/204Pb, Ka‘ena and Wai‘alu samples Gaffney et al., 2004; Garcia et al., 2010). These of Ka‘ena Point, coincident with a peak in dike are more similar to volcanoes of the “Loa-trend” secondary lava compositions require an isotopi- intensity in that location (Stearns and Vaksvik, on the island of Hawai‘i, whereas Wai‘anae Vol- cally depleted source component that does not 1935; Stearns, 1939; Zbinden and Sinton, 1988). cano might be classifi ed as a “Kea-type” (Stille appear to be a major contributor to Hawaiian Two smaller anomalies are present on et al., 1986; Abouchami et al., 2005). However, shield volcanoes. Ka‘ena Ridge, both showing peak amplitudes unlike isotopic data for Hawai‘i Island vol- ~15%–20% of the main Wai‘anae anomaly. The canoes, Wai‘anae and especially Ka‘ena isotope Gravity Anomalies larger of the two is centered 15 km northwest ratio variations are highly oblique to values of of Ka‘ena Point, near 21°42′N, 158°22′W; a constant 208Pb*/206Pb* (Galer and O’Nions, Gravity data were collected along 75–140-km- smaller anomaly is located just to the east of 1985), a feature these volcanoes have in com- long survey lines spaced 4–6 km apart, and the eastern bathymetric high on Ka‘ena Ridge mon with Ko‘olau and West Moloka‘i (Tanaka oriented ~N10E, approximately orthogonal (Fig. 10). According to the inversion results et al., 2002; Xu et al., 2007). to the strike of Ka‘ena Ridge during cruises of Flinders et al. (2013), the eastern anomaly

12 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

7 One of the dated Ka‘ena samples is a transi- mugearite tional Ka‘ena B basalt (KM 6–1, age 3.6 Ma; Wai‘anae postshield Table 5), which represents a late stage of Ka‘ena 6 alkalic evolution, close in time to the sub marine to sub- hawaiite transitional aerial transition of Ka‘ena Ridge. The ages of Ka‘ena A sample KM 9–1 and Ka‘ena B sample 5 tholeiitic KM 6–1 are indistinguishable (Table 5), sup- alkali basalt porting the interpretation that these two chemi- cal types were being erupted at about the same 4 + + + + Ko‘olau

O (wt%) time. Geological observations suggest that the 2 + + subaerial section of Ka‘ena Ridge is relatively ++ 3 + + thin, and presumably short-lived. We dated + + ++ ++ O + K + + + + + two samples by K/Ar that likely represent 2 + + + + + + sub aerially erupted lavas (vesicular pāhoehoe + + + Na 2 + + + sample KM 9–4, and massive basalt J551–11); + both yielded ages younger than 3 Ma (Table 5; secondary basalt T-325 Fig. 11). Although these ages could be too Ka‘ena B 1 J305/306 + young, presumably owing to Ar loss, their rela- Ka‘ena A + Wai‘anae tively young ages leave the end of volcanism on Wai‘alu shield Ka‘ena Ridge unresolved. Despite this uncer- 0 tainty, it is apparent that Ka‘ena Ridge was mak- 44 46 48 50 52 54 56 ing the transition from tholeiitic to alkalic com- SiO (wt%) 2 position ~3.6 m.y. ago. In contrast, the oldest alkalic lavas of Wai‘anae Volcano were erupted Figure 5. Total alkalis vs. SiO2 (wt%) for sample data reported in Tables 1 and 2 (glass data ca. 3.2 Ma, within the late-shield Kamaile‘unu by electron microprobe are shown with smaller symbols than X-ray fl uorescence [XRF] data Member in a section just above the Mammoth on whole rocks) compared to whole-rock data from Wai‘anae Volcano (shield samples—small magnetic polarity reversal. gray dots; dash-dot fi eld encloses Wai‘anae postshield data; Wai‘anae data from Sinton, 1986; Moore and Clague (2004) showed that the Zbinden and Sinton, 1988; Presley et al., 1997; Guillou et al., 2000; Sherrod et al., 2007; van thickness of Mn-Fe rinds on seafl oor samples der Zander et al., 2010), data for J2 dives 305 and 306 (+) from Greene et al. (2010), and for can be used as an indicator of age. Although Tiburon dive T-325 (inverted triangles) from Coombs et al. (2004). Dotted fi eld encloses data complications arise from postdepositional from the Ko‘olau Volcano from Frey et al. (1994), Jackson et al. (1999), Tanaka et al. (2002), breakage, landslides, and intermittent burial, and Haskins and Garcia (2004). Dashed lines are parallel to the line dividing alkalic from tho- the maximum thickness on individual samples leiitic compositions of Macdonald (1968); in this paper, we defi ne transitional basalts as those corresponds with radiometric ages, using a crust samples with A.I. = 0 ± 0.5, where A.I. = alkalinity index of Rhodes and Vollinger (2004). Other growth rate of 2.5 mm/m.y. (Moore and Clague, classifi cation labels are from Le Bas et al. (1986). Analytical uncertainty for whole-rock data is 2004). We measured average and maximum shown by the dark cross in the upper-left corner. All Wai‘alu Ridge and all but one Ka‘ena A Mn-Fe rind thicknesses on 48 volcanic and samples are tholeiitic basalts; Ka‘ena B samples include transitional and alkalic basalts and 24 hyaloclastite samples from our collections. one hawaiite; secondary samples range from alkali basalt to basanite. Despite considerable variation, most volcanic samples have maximum rind thickness in the range 7–12 mm (supplementary material, Fig. region has slightly higher average density Age of Ka‘ena Ridge S1 [see footnote 1]). If rind growth occurred at (~2.96 g/cm3) than the smaller, western anom- 2.5 mm/m.y., these values suggest ages of 2.8– aly region (~2.88 g/cm3). Neither qualifi es as a Greene et al. (2010) reported ages of 4.9– 4.8 Ma, spanning the range of radiometric ages signifi cant intrusive core complex. Instead, the 3.7 Ma for samples from the western edge of obtained from Ka‘ena Ridge (Fig. 11). subdued density contrasts can be accounted for Ka‘ena Ridge. We dated four additional samples by >60% dikes with average density of 2.95 from Ka‘ena Ridge (Table 5). The combined DISCUSSION g/cm3, or limited (~5%–10%) accumulation of data show that Ka‘ena Ridge samples clus- olivine in the subsurface. ter mainly between 3.5 and 4.9 Ma (Fig. 11), Geochemical and Geochronological The principal features of the gravity data for extending ~1 m.y. older than the oldest dated Evidence for Ka‘ena Volcano Ka‘ena Ridge are the northward offset from the Wai‘anae lavas (Guillou et al., 2000). The 2.85– main Wai‘anae anomaly and the presence of two 3.9 Ma age of Wai‘anae Volcano is well known The interlayering of chemical types at distinct anomalies with overall low magnitude. from more than 60 radiometric ages (summa- Ka‘ena Ridge is typical of many Hawaiian vol- The volume and magnitude of the Ka‘ena anom- rized in Sherrod et al., 2007), many of which canoes, in which an early shield stage erupts alies are comparable to those associated with the are constrained further by frequent magnetic exclusively tholeiitic basalts, followed by a Kīlauea east rift zone, Hāna Ridge, Lāna‘i, and reversals that occur between 3 and 4 Ma (e.g., later shield stage that marks the transition to Hualālai Volcanoes, and they are much smaller Doell and Dalrymple, 1973; Presley et al., 1997; alkalic volcanism (Macdonald, 1968; Sinton, in both volume and magnitude than those asso- Guillou et al., 2000). These results suggest that 1986, 2005; Zbinden and Sinton, 1988; Sherrod ciated with most other major Hawaiian shield volcanism on Ka‘ena Ridge predated that of the et al., 2007; Clague and Sherrod, 2014). Tho- volcanoes (Flinders et al., 2013). subaerial Wai‘anae Volcano by up to 1 m.y. leiitic and alkalic lavas are interbedded within

Geological Society of America Bulletin, Month/Month 2014 13 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al.

A + B 54 + 14 + + + + + 52 + + + + + + + + + + + 13 + + + + + ++ ++ + + + 50 +++

(wt%) +

2 + 12 + + ++ + + + + +

SiO + 48 FeO* (wt%) ++ + + + ++ + + 11 + ++ + + + + + + + + ++ + + + + 46 10 ++ +

C D secondary 1.2 T-325 Wai‘anae postshield Ka‘ena B 60 J305/306 + Ka‘ena A 1.0 Wai‘anae 50 Wai‘alu shield 0.8 40 (wt%) 5 +

O 0.6 2

Nb (ppm) 30 P +++ + + + + + 0.4 ++ + + + + 20 ++ +++++ + ++ + + + + + ++ ++ + 0.2 + + + + + ++ + ++ + 10 + + +++ ++ + + + + + +++++ Ko‘olau + + + + + ++ + +

E F (3.4) 1.8 0.9

+ 0.8 3 + 1.4 + + + O 2 +++ + + + +++++ + 0.7 + ++ + ++ + + + + + + + Ba/V 1.0 ++++ + 0.6 +

CaO/Al + +++ + + ++ + 0.5 0.6 + + + ++ + +++ + + +++ 0.4 + + + + + 0.2 + + ++ +

3 5 7 9 11 13 15 3 5 7 9 11 13 15 MgO (wt%) MgO (wt%)

Figure 6. (A–F) Compositional variation vs. MgO for whole-rock samples; all data and fi eld symbols are same as in Figure 5; error on Nb data is less than 1 ppm (D). Ka‘ena A and Wai‘alu Ridge samples are characterized by relatively high SiO2, and low FeO*, P2O5, and Nb, broadly similar to samples from the Ko‘olau Volcano, but with even lower Ba/V (F). Ka‘ena B compositions have generally lower SiO2 and higher FeO*,

Nb, and P2O5 than Ka‘ena A. Samples from Tiburon dive T-325 include both Ka‘ena A and Ka‘ena B compositions. See text for discussion.

the late-shield stage of Wai‘anae Volcano 2007), although alkalic basalts become increas- (data compilation in Sherrod et al., 2007). The (Kamaile‘unu Member of the Wai‘anae vol- ingly common up section. In addition to alkalic late-shield succession at Wai‘anae is overlain canics; Sinton, 1986; Zbinden and Sinton, 1988) basalt, minor hawaiite also is found in the late- by a postshield sequence ranging in composi- and many other Hawaiian volcanoes, includ- shield stages of the Wai‘anae (Kamaile‘unu), tion from alkalic basalt to mugearite (Pālehua ing Kohala, Mauna Kea, East Maui, and East Kohala (Pololū), Mauna Kea (Hāmākua), West and Kolekole Members of the Wai‘anae vol- Moloka‘i (Macdonald, 1968; Sherrod et al., Maui (Wailuku), and Kaho‘olawe Volcanoes canics; Presley et al., 1997).

14 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

Figure 7. Chondrite-normalized, lantha- 100 nide rare earth element (REE) variations A for Ka‘ena and Wai‘alu Ridge and sec- Ka‘ena A ondary lava samples. (A) REE patterns for + Wai‘alu Ka‘ena A (solid lines) and Wai‘alu Ridge (short dash) samples. Sample J556–3 has a slightly steeper slope (higher La/Yb) and J556–9 has a lower slope (lower La/Yb) than J556-9 other Ka‘ena A samples. Light-gray-shaded fi eld shows data of Coombs et al. (2004) REE/Chondrite J556-3 and Greene et al. (2010) that we interpret 10 to have Ka‘ena A compositions. Our data include samples with lower MgO and cor- respondingly higher total REE values than previously reported. (B) Ka‘ena B samples. La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Dark-gray-shaded fi eld shows the range of Ka‘ena A and Wai‘alu Ridge samples from B panel A. Ka‘ena B samples have similar Ka‘ena B heavy REE values to Ka‘ena A but are en- riched in light REEs. (C) REE patterns for secondary lava samples (shaded fi eld same 100 as in panel B). Secondary lava samples are depleted in heavy REEs compared to Ka‘ena A and B, but with light REEs comparable Ka‘ena A to those in Ka‘ena B. Alkali basalt sample + Wai‘alu J381–8 has a lower overall slope than the other basanite secondary lava samples. REE/Chondrite

Although the volcanic successions in both 10 Ka‘ena Ridge and Wai‘anae Volcano can be characterized in terms of early- and late-shield La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu evolution, there are signifi cant differences between them, including the apparent lack of C a well-developed postshield stage at Ka‘ena secondary Ridge (Fig. 12). Most of the samples from Ka‘ena and Wai‘alu Ridges have tholeiitic, 100 Ka‘ena A compositions, which tend to have higher SiO2 and lower FeO* and Ba/V than is typical of Wai‘anae tholeiites. In this regard, J381-8 Ka‘ena A lavas are more similar to those of the Ko‘olau Volcano. The youngest parts of Ka‘ena Ridge are dominated by the eruption of lavas with Ka‘ena B chemical characteristics, which Ka‘ena A we interpret to have formed by lower extents REE/Chondrite + Wai‘alu of melting from mantle sources that lie along a mixing trend similar in radiogenic isotope com- 10 positions to that of Ka‘ena A (Fig. 9). Chemi- cally similar Wai‘alu Ridge samples also lie on La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu this mixing line. The Ka‘ena-Wai‘alu mixing 208 204 trend has higher Pb/ Pb and lower εNd for a given 206Pb/204Pb compared to most Wai‘anae samples (Fig. 9). Thus, Ka‘ena and Wai‘alu different evolutionary histories. The transition of a conundrum, as volcanoes in the Hawaiian Ridges appear to constitute a single magmatic from Ka‘ena A to Ka‘ena B occurred just as chain generally are younger to the southeast. system that is chemically distinct from that of Ka‘ena Volcano was emerging above sea level However, several lines of reasoning indicate that Wai‘anae Volcano. We refer to this volcanic sys- 3.5–3.6 Ma, i.e., several hundred thousand years Wai‘anae is a younger magmatic center than tem as Ka‘ena Volcano. after Wai‘anae was already a large, subaerial Ka‘ena, despite its earlier emergence. Ka‘ena and Wai‘anae are chemically dis- volcano (Fig. 12). The apparent emergence of The oldest alkalic lavas of Ka‘ena are ca. tinct, but more importantly, they have very Ka‘ena later than Wai‘anae represents somewhat 3.6 Ma, whereas the oldest alkalic volcanism of

Geological Society of America Bulletin, Month/Month 2014 15 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al.

Samples collected from the deep, western secondary slopes of Ka‘ena Ridge have been referred to 2000 Ka‘ena B as West Ka‘ena (Greene et al., 2010). Their Ka‘ena A location on the western end of Ka‘ena Ridge Wai‘alu is consistent with formation along rift zones of Ka‘ena Ka‘ena Volcano. They have Ka‘ena A magma 1500 Slide composition, as defi ned here. We therefore consider West Ka‘ena to simply represent dis- tal eruptions from the larger Ka‘ena Volcano that is responsible for the main construction of

S (ppm) 1000 Ka‘ena Ridge.

Structure and Morphology of Ka‘ena Volcano 500 degassed Ka‘ena Ridge forms a general prolongation of the topographic axis of the Wai‘anae mountain range on the west side of the island of O‘ahu. However, topography of the Wai‘anae Range 345results dominantly from erosion, particularly in response to major mass wasting (Moore et al., Na2O + K2O (wt%) 1989; Presley et al., 1997). Lava fl ows that form Ka‘ena Point dip 5–10°N, confi rming that the Figure 8. Sulfur contents in glass samples from Wai‘alu Ridge, Ka‘ena Ridge, Ka‘ena slide, western tip of O‘ahu is an erosional remnant of and secondary lava fl ows plotted against total alkali contents. Ka‘ena slide samples from the north fl ank of Wai‘anae Volcano, and that the this study (Table 1) are shaded; data from Tiburon dives T-327 and T-328 are shown as open constructional volcanic axis must lie to the south squares. S contents above 800 ppm generally correlate with total alkalis and FeO* (not of the topographic axis. The northwest rift zone shown). The lack of correlation for S contents <500 ppm refl ects variable S degassing at low of Wai‘anae Volcano is defi ned by subparallel pressure. Degassed samples likely erupted subaerially or in shallow water (<1000 m). dikes striking ~N60°W, with a strong concentra- tion in dike intensity centered ~7 km southeast of Ka‘ena Point (Stearns, 1939; Zbinden and Wai‘anae Volcano is well constrained by radio- In contrast, Ka‘ena appears to have remained Sinton, 1988). metric dating and magnetic reversal stratigra- submarine for ~1.5 m.y. (Fig. 12), an unusually Topographic lineaments associated with phy to be ca. 3.2 Ma. Thus, the evolution from protracted history of submarine growth (e.g., Ka‘ena and Wai‘alu Ridges suggest two major tholeiitic to mixed tholeiitic-alkalic volcanism Lipman and Calvert, 2013), and only emerged rift zones that might be associated with mag- occurred ~0.4 m.y. earlier on Ka‘ena than on very late in its life span. As the oldest edifi ce matic centers coincident with the Ka‘ena grav- Wai‘anae (Fig. 12). In terms of chemical evolu- in the O‘ahu region, Ka‘ena initiated in deep ity anomalies (Fig. 3). Notably, the depth profi le tion, Ka‘ena preceded Wai‘anae, consistent with water in the Kaua‘i Channel, whereas Wai‘anae of Wai‘alu Ridge is consistent with it being a an overall southeastward age progression. largely grew on the fl ank of Ka‘ena. Ka‘ena submarine rift zone of Ka‘ena Volcano (Fig. 2), The oldest ages for Ka‘ena are almost 5 Ma bears some similarities to the evolution of but it is likely too shallow to be an extension (Figs. 11 and 12), whereas the oldest reliable Māhukona volcano, which is likely to have been of Wai‘anae Volcano, lying >50 km to the dates for the subaerial Wai‘anae Volcano are the fi rst volcano of the Hawai‘i Island complex southeast. It also has an orientation unlike any 3.9 Ma (Guillou et al., 2000). The submarine (Moore and Clague, 1992) and appears to have known for Wai‘anae Volcano. The J2–560 dike inception of Wai‘anae volcanic activity and its begun erupting transitional basalts shortly after complex indicates a major structure striking SE, emergence above sea level are unknown, pri- emergence (Clague and Moore, 1991; Clague whereas rare NNE-striking dikes in J2–551 are marily because deep submarine Wai‘anae sam- and Calvert, 2009). Thus, isolated volcanoes parallel to the topographic ridge south of the ples have not been found; the offshore western initiating in deep water might experience longer eastern topographic high on Ka‘ena Ridge. Not side of Wai‘anae is covered with slump deposits, submarine growth histories than volcanoes initi- all of these dikes were necessarily active simul- and the eastern side is buried beneath Ko‘olau ating on the shallow fl anks of earlier ones. taneously. Dikes from J2–560 have Ka‘ena A lava fl ows. Although no single growth history Although samples from Ka‘ena bear some compositions, indicating that they represent characterizes all Hawaiian volcanoes, (e.g., Lip- major- and trace-element similarities to the relatively early shield-stage magmatism. The man and Calvert, 2013), DePaolo and Stolper Ko‘olau Volcano, Ko‘olau is ~1 m.y. younger eastern summit, and probably both topographic (1996) argued that the early submarine history than Wai‘anae (for review, see Sherrod et al., summits on Ka‘ena Ridge consists of late-form- should be characterized by high growth rates. 2007), and therefore at least 1.5 m.y. younger ing, subaerial lavas. Estimates of the time from birth to emergence than Ka‘ena. Taken together, the magmatic evo- The residual gravity anomalies that we for Kohala, Mauna Kea, and Kīlauea Volcanoes lution of Ka‘ena appears to be both temporally associate with Ka‘ena Volcano have lower on Hawai‘i Island are in the range 250–400 k.y. and chemically distinct from that of other O‘ahu anomaly magnitude (~10 mGal) than those of (Moore and Clague, 1992; Lipman and Calvert, volcanoes, further supporting the interpretation the Wai‘anae and Ko‘olau Volcanoes, both of 2013). Assuming a similar growth history for that it represents a major precursor volcano to which have residual gravity anomaly magni- Wai‘anae yields an inception age of ca. 4.5 Ma. the island of O‘ahu. tudes >50 mGal, approximately equivalent to

16 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

0.7042 AB + Ko + 38.2 0.7040 + + 38.1 + 0.7038 + + ++ ++ + W

Pb 38.0

Sr + Ws + ++ Ka‘ena trend 86 + 204 0.7036 Ko W Sr/ 37.9 Pb/ 87 + + 208 0.7034 37.8 secondary + ++ Ws + 0.7032 37.7 ++ + + + +

C 17.8 18.0 18.2 18.4 18.6 9 206Pb/204Pb

8 D + Ws 7 0.97 + W + +

Nd 6 + Ko + + ++ ε

Pb* 0.96 + + + + + + + ++ + + 206 5 Ws ++ + + + +

Pb*/ 0.95

4 + + 208 W Ko 0.94 3

17.8 18.0 18.2 18.4 18.6 4.5 5.5ε 6.5 7.5 8.5 206Pb/204Pb Nd

Figure 9. Isotopic variations for Ka‘ena and Wai‘alu Ridge and secondary volcanic samples. Small open diamonds are secondary alkalic data from Greene et al. (2010); other symbols are same as in Figure 5. Also shown are isotopic data for Wai‘anae Volcano (van der Zander, 2010; Table 4 herein) and fi elds for Ko‘olau Volcano samples (dashed fi eld labeled Ko; data from Tanaka et al., 2002; Abouchami et al.,

2005; Fekiacova et al., 2007). Most Wai‘anae data plot within the solid fi eld enclosed by a solid line (W) and have slightly higher εNd and lower 208Pb/204Pb than Ka‘ena and Wai‘alu Ridge samples (shaded symbols; trends shown by thick gray lines). Some early-shield and high- silica Wai‘anae samples (Ws; van der Zander et al., 2010) lie along trends with higher 208Pb/204Pb, similar to those from Ka‘ena and Wai‘alu Ridges. All errors are smaller than the symbols.

300 mGal complete Bouguer gravity anomaly of dense, olivine-rich cumulates. Smaller grav- composition toward the end of its lifetime. (Strange et al., 1965; Flinders et al., 2013). ity anomalies could refl ect magmatism that is Thus, if two separate volcanoes were present Although the Ka‘ena anomalies are smaller preferentially distributed into rift zones, or par- at Ka‘ena Ridge, they erupted magmas derived than those of many Hawaiian volcanoes, they titioned into extrusion over intrusion. from similar mantle sources and with parallel are comparable in amplitude to those of Lāna‘i The presence of multiple rift zones and two chemical evolution over similar life spans. In and Hualālai, and they are substantially larger separate centers within a broad gravity anomaly this regard, the two-volcano hypothesis does than those of Māhukona and Penguin Bank (see could be interpreted in terms of more than one not meet the combined criteria for independent Fig. 3 in Flinders et al., 2013). The volume and volcano on Ka‘ena Ridge (Flinders et al., 2013). structure, magmatic evolution, and age. We amplitude of gravity anomalies associated with Multiple gravity anomaly centers can character- therefore prefer the interpretation that a single, Hawaiian volcanoes tend to correlate with vol- ize individual volcanoes, however, as shown by spatially extensive major Hawaiian shield vol- cano volume, although some volcanoes with the presence of a secondary anomaly near Pu‘u cano built Ka‘ena and Wai‘alu Ridges. estimated volumes >30,000 km3 have relatively ‘Ō‘ō on Kīlauea’s east rift zone (Flinders et al., There are no identifi able caldera structures small density anomalies (Flinders et al., 2013). 2013). Available geochemical, geochronologi- on Ka‘ena Ridge, although the area of the Gravity anomalies should be maximized in cal, and morphological data from Ka‘ena and larger gravity anomaly is now buried beneath large volcanoes with spatially persistent mag- Wai‘alu Ridges are consistent with a single Wai‘anae lava fl ows (Fig. 3). The lack of a topo- matic conduits containing signifi cant volumes magmatic system, albeit one that evolved in graphic caldera is consistent with our interpreta-

Geological Society of America Bulletin, Month/Month 2014 17 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al.

tion that Ka‘ena evolved to the late-shield stage of evolution, which Macdonald (1968) charac- terized as a time of caldera fi lling. Although it is now recognized that calderas can form and fi ll N repeatedly during the shield stage (e.g., Clague 0 and Sherrod, 2014), the late-shield stage likely 10 represents the time of ultimate caldera fi lling, after which the lack of surface calderas corre- 10 21°40′N sponds to the disappearance of shallow magma chambers (Sinton, 2005). In terms of vol canic evolution, Ka‘ena bears some similarity to Kaho‘olawe, a 460-m-high (present height) O‘ahu shield volcano, where some of the postcaldera topographic peaks are composed of tholeiitic 30 lavas (Fodor et al., 1992). 10 40 The thickness of the subaerial section on 5050 Ka‘ena Ridge suggests that the volcano likely

20 never reached more than ~1000 m above sea level. However, Ka‘ena was built on the abys- 0 sal ocean fl oor and is ~4 km high. Multiple rift zones distributed magma over a broad area, 21°20′N which contributed to the unusually long sub- 01020 marine growth history, relatively low-amplitude kilometers gravity anomalies, and limited height above sea level, compared to many volcanoes in the Hawaiian chain. 158°30′W 158°00′W Evidence for Tilting of Ka‘ena Ridge Figure 10. Residual gravity data (mGal), shown as smoothed, dashed contours. A nearly circular region of maximum residual gravity (>40 mGal) coincides with the geologically Stratigraphic horizons indicative of former defi ned Wai‘anae caldera. The Wai‘anae anomaly is elongate parallel to dike strikes on land sea levels vary systematically in depth around and crosses the shoreline south of Ka‘ena Point. Two small, ~10 mGal, anomalies are pres- Ka‘ena and Wai‘alu Ridges. The top of the sub- ent on Ka‘ena Ridge. marine pillow section deepens to the south by at least 800 m over a distance of 40 km, from ~1350 mbsl on Wai‘alu Ridge and ~1850 m on dive J2–556, to between 1600 m and 2400 m in KM06 dredge 9, and ~2200 m on the south side 5 of Ka‘ena Ridge. The base of the subaerial sec- tion is less well defi ned because of signifi cant sediment redistribution and sparse outcrops in 4 the upper parts of Ka‘ena Ridge, but vesicular pāhoehoe lava was recovered in KM dredges 6 and 9, and outcrops of subaerial lava were

3 observed at ~1650 m depth in both dives J2–551 { ages may be and J2–560. The tops of the steep slopes domi- too young nated by hyaloclastite likely represent the emer- Age (Ma) gent subaerial transition, and this topographic 2 break occurs at least 500 m deeper on the south side of Ka‘ena Ridge than on the northwest end near dive J2–556. On Wai‘alu Ridge, the slope 1 secondary { break is another ~100 m shallower than on the alkalic west end of Ka‘ena Ridge, consistent with the 1.5°–2° southerly dip indicated by observa- 0 tions on opposing sides of Ka‘ena Ridge, and Sequential order by decreasing age as expected if the shorelines on Ka‘ena and Wai‘alu were synchronous. These geometric Figure 11. Compilation of age data for the region of this study. New K/Ar data reported here relations provide additional support for the (triangles; darker shading is Ka‘ena B sample KM6–1; lighter shading represents Ka‘ena A interpretation, based on chemical and iso topic samples) are compared to data of Greene et al. (2010) for Ka‘ena A (circles) and secondary similarity, that Wai‘alu Ridge is part of the (diamonds) lava samples; all Greene et al. (2010) data are 40Ar/39Ar plateau ages. greater Ka‘ena Volcano.

18 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

Age (Ma) landslides around the Hawaiian Islands does Ka‘ena Wai‘anae not involve volcano-scale differential vertical motions, and the volume of secondary volca- nism in the Kaua‘i Channel to the southwest is insuffi cient to provide much of a load. The 3.0 postshield detailed subsidence history of the Wai‘anae and Ko‘olau Volcanoes is very poorly known, shield however, at least partly because critical parts first alkalic magmas of the offshore regions have been signifi cantly modifi ed by landsliding. New observations of paleoshorelines on these and other Hawaiian first transitional lavas volcanoes are necessary to better understand the timing and orientation of differential subsidence 3.5 subaerial along the Hawaiian Ridge.

Constraints on Age and Sources of Ka‘ena and Wai‘anae Landslide Deposits

The identifi cation of compositional variations oldest reliably dated rock within Ka‘ena Volcano allows us to re assess the 4.0 (subaerial tholeiite) analytical data of Coombs et al. (2004) in the context of the timing and composition of nearby landslide deposits. Coombs et al. (2004) col- ? lected samples from two ROV Tiburon dives in submarine ? the Ka‘ena slide, and we obtained glass from KM dredge 2 (Fig. 1; Table 1). We recognize both Ka‘ena A and Ka‘ena B chemical types in glass data from these sites (Fig. 13), including some transitional basalts with low SiO and high 4.5 2 FeO* and P2O5, typical of Ka‘ena B. Ka‘ena A

glasses from dive T-327 have high SiO2 and low subaerial alkalic FeO*, which distinguish them from Wai‘anae compositions. Whole-rock data for samples mixed from dive J2–251 (Fig. 1) also have Ka‘ena A submarine tholeiitic composition (Table 2). Thus, samples from the four sample locations in the Ka‘ena slide (T-327, 5.0 T-328, J2–251, KM dredge 2) have composi- tions consistent with derivation from Ka‘ena Volcano, and they require no contribution from Wai‘anae. Because Ka‘ena A compositions are Figure 12. Geologic and magmatic evolution of Ka‘ena and Wai‘anae Volcanoes. Individual diffi cult to distinguish from Ko‘olau, we cannot radiometric ages for Ka‘ena (triangles; same data as shown in Fig. 11) are plotted to the preclude some contribution from that volcano. left of the Ka‘ena column; shaded symbol is Ka‘ena B sample KM6–1. One of the two ages The presence of Ka‘ena B compositions in thought to be too young from Figure 11 is shown with a dashed outline; the other “young” Ka‘ena slide deposits indicates that at least subaerial sample and all secondary lava ages of Figure 11 plot off the scale of the diagram. some of the sliding occurred late in the history Wai‘anae age data from Presley et al. (1997), Laj et al. (1999), and Guillou et al. (2000) of Ka‘ena Volcano, subsequent to 3.6 Ma. Addi- (squares) are plotted next to the Wai‘anae column, where shaded symbols denote alkalic tionally, most glass samples from the Ka‘ena samples. Most of Ka‘ena history is submarine, with emergence coinciding with the late- slide have low sulfur contents (Table 1; Fig. 8), shield transition. In contrast, Wai‘anae emerged much earlier, possibly because Wai‘anae further supporting the conclusion that much of initiated on the fl anks of the preexisting Ka‘ena Volcano. The late-shield transition to alkalic the deposit was derived from high-level lava volcanism occurs at least 0.4 m.y. later on Wai‘anae than on Ka‘ena. fl ows emplaced after Ka‘ena Volcano had grown to shallow water depths. Wai‘anae was an active subaerial volcano at this time, although we have Correlations among Quaternary reef terraces which they attributed to loading by the next not recognized any Wai‘anae-derived debris in (Moore and Campbell, 1987) indicate tilting in seamount to the south. In contrast, the tilting the Ka‘ena slide, presumably because Wai‘anae a direction subparallel to the Hawaiian Ridge, of Ka‘ena Ridge is predominantly in a south- was mainly built on the east fl ank of Ka‘ena, and attributed to recent loading by volcanoes lying erly direction, at high angle to the strike of not affected by sliding of Ka‘ena’s north fl ank. to the ESE. Lonsdale et al. (1993) identifi ed the Hawaiian Ridge, and therefore unlikely to In contrast, as noted by Coombs et al. (2004), 2°–3° of SSE tilt of the summit platform of be a fl exural response to down-ridge loading. samples from the Wai‘anae slump can mostly, or Suizei Guyot in the Emperor Seamount Chain, Current understanding of the large submarine perhaps entirely, be ascribed to derivation from

Geological Society of America Bulletin, Month/Month 2014 19 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al.

Figure 13. Glass chemical data from this study (triangles, shaded squares, see legend) A compared to glass data from the Ka‘ena and 55 Wai‘anae landslide deposits (open squares and circles, respectively) from Coombs et al. 54 (2004); also shown are the fi elds for selvage glasses from Wai‘anae dikes (dashed line; 53 data of Zbinden and Sinton, 1988). Esti- mated errors for Hawai‘i microprobe data 52 are shown by bold crosses in lower-left cor- (wt%) 2 ner of each panel. All data from the Ka‘ena 51 slide can be classified as either Ka‘ena A SiO 50 (high SiO2, low FeO*) or Ka‘ena B (low

SiO2, high FeO*, high P2O5). In contrast, most data from the Wai‘anae slump (small 49 circles) have intermediate SiO2, within the range of data for Wai‘anae dikes. 48

B the shield stage of Wai‘anae Volcano. Some 15 glasses with high SiO2 and low FeO* from the K205 collection of Coombs et al. (2004) (Fig. 13) might represent a minor contribution 14 of Ka‘ena A compositions, but high compo- Wai‘anae dikes sitional variability with considerable overlap 13 makes this interpretation tenuous. Tiburon dive T-326 (Fig. 1) is overwhelmingly dominated by compositions similar to Wai‘anae shield lavas 12 (Coombs et al., 2004). One sample (T-326,

PC61G) has a composition broadly similar FeO* (wt%) 11 to our Ka‘ena B, although again overlap with Wai‘anae Volcano compositions at this range of MgO makes interpretation equivocal. 10 Despite the lack of identifi able Ka‘ena mate- rial in the sparse sampling within the Wai‘anae 9 slump, steep headwall scarps cross the bound- ary between Ka‘ena and Wai‘anae Volcanoes (Fig. 3). It is therefore inescapable that some C Ka‘ena A portion of the Wai‘anae slump deposit must be 0.7 Ka‘ena B derived from Ka‘ena Volcano. The complete Wai‘alu history of the Wai‘anae slump is unknown. Presley et al. (1997) showed that profound ero- 0.6 Ka‘ena Slide sion separating the Pālehua and Kolekole post- Wai‘anae Slump shield members of Wai‘anae Volcano is slightly 0.5 younger than 3 Ma and coupled to major dis-

ruption of the Wai‘anae Volcano magmatic (wt%) 5 system; they suggested that this period might

O 0.4 be associated with major mass wasting late in 2 the history of Wai‘anae Volcano. Coombs et al. P (2004) cited the lack of postshield compositions 0.3 in the slump deposit as evidence that most of the Wai‘anae slump occurred during the shield 0.2 stage. However, Hawaiian postshield sequences tend to thin rapidly toward the coast, whereas the headwall scarps for the Wai‘anae landslide are mostly >10 km offshore, far from the known 456789 outcrop area of Wai‘anae postshield lava fl ows. MgO (wt%) Nonetheless, slump deposits are generally con- sidered to form from multiple slides, and there

20 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

are three distinct scarps on the south side of the volume of the Ka‘ena part of the offshore ridges Ko‘olau Volcano (for a review, see Sherrod et al., Ka‘ena-Wai‘anae landmass (Fig. 3). We suggest to be 14.0 × 103 km3, with an additional Ka‘ena- 2007) are close to the end of Wai‘anae postshield that early failure of the Ka‘ena and Wai‘anae derived volume of 3.3 × 103 km3 in the Ka‘ena activity, suggesting a shorter period of concur- edifi ces might have occurred in the northern slide (Table 6). Thus, compared to the values of rent eruption for these volcanoes. part of the slumping area, producing the shal- Robinson and Eakins (2006), the Wai‘anae Vol- In order to estimate volumes that better low scarps rising to ~1000 m. Later slumping cano volume is reduced by 17.3 × 103 km3, to account for overlapping volcano edifi ces, we occurring in deeper water spread a blanket of yield a Wai‘anae volume of 36.2 × 103 km3. Ito project surface contacts between adjacent vol- Wai‘anae shield debris over the earlier depos- et al. (2013) determined a volume of 36.8 × 103 canoes along a constant slope down to the top its. This scenario, although speculative, is con- km3 for Wai‘anae Volcano, also assuming verti- of the Pacifi c plate, which, following Robinson sistent with the presence of multiple landslide cal contacts and excluding Ka‘ena Ridge from and Eakins (2006), we defi ne as the top of the scars and the lack of recovered Ka‘ena debris in their calculation. oceanic crust (Watts and ten Brink, 1989) plus the surfi cial Wai‘anae slump deposits. Although the recognition of a separate vol- a 500-m-thick sediment layer. Volumes are then cano at Ka‘ena Ridge suggests that Wai‘anae is calculated for the enclosing volcanic edifi ces. Assembling O‘ahu: Geometric Constraints much smaller than previously proposed, these Because of the uncertainty that surrounds the and Volume Estimates for O‘ahu Volcanoes calculations do not take into account the vol- choice of slopes, we have calculated the buried ume of Ka‘ena Volcano that underlies Wai‘anae edifi ce volumes using both 10° and 30°, which Volume calculations for Hawaiian volcanoes Volcano, or the volume of Wai‘anae Volcano presumably covers the range that might accrue are subject to considerable uncertainty because that underlies Ko‘olau. The slopes on subaerial from combinations of subaerial and submarine it is diffi cult to know precisely the extent of Hawaiian shield volcanoes generally range slopes, interfi ngering between concurrent vol- older edifi ces beneath younger volcanoes built between ~5° and 15° (e.g., Stearns, 1939, 1946; canism, and increased loading by younger edi- on their fl anks. Robinson and Eakins (2006) Macdonald et al., 1983), except for local varia- fi ces. Values calculated by this method are given argued that there simply is too little information tions related mainly to faulting. Eruption of dif- in Table 6. to account for nonvertical, nonplanar contact ferentiated alkalic lavas during postshield-stage We estimate the amount of Ka‘ena Volcano relations and invoked vertical contacts between volcanism can lead locally to slopes as steep underlying Wai‘anae Volcano to be in the range adjacent volcanoes in their review of volcano as ~20° (Mark and Moore, 1987). Submarine 1.5–4.4 × 103 km3. Similarly, the amount of volumes along the young end of the Hawaiian slopes also can be steeper (e.g., Mark and Moore, Wai‘anae Volcano that underlies Ko‘olau is Ridge. They calculated edifi ce volumes to the 1987), especially in areas of signifi cant hyalo- estimated to be 2.7–8.4 × 103 km3. We do not top of the oceanic crust beneath Hawaiian vol- clastite accumulation and near landslide head- recognize any contribution from Wai‘anae to canoes (Watts and ten Brink, 1989), a signifi - walls. Furthermore, when one volcano is built the Ka‘ena slide (see previous), but the distribu- cant improvement to volume calculations made on a preexisting edifi ce, the basal contact could tion of scarps around the south side of Ka‘ena before the magnitude of subsidence along the be affected by additional loading and increased Ridge requires some contribution from Ka‘ena Hawaiian chain was known (e.g., Bargar and subsidence. Finally, concurrently active, adja- to the Wai‘anae slump. We have therefore used Jackson, 1974). cent volcanoes will have interfi ngering contacts. the same projection methods to evaluate the However, the assumption of vertical contacts Existing data suggest that Ka‘ena and Wai‘anae relative contribution of Ka‘ena to the Wai‘anae between adjacent volcanoes can lead to large were both active 4.0–3.5 Ma (Fig. 12), and possi- slump, with results that range from ~10% to errors in the calculation of volcano volumes bly longer, and that Wai‘anae continued to erupt 30% (Table 6). (DePaolo and Stolper, 1996; Lipman et al., a signifi cant volume for ~0.5 m.y. after Ka‘ena Our estimate for the volume of Ka‘ena Vol- 2006; Lipman and Calvert, 2013). Volumes production waned. The oldest available ages for cano ranges from 20 to 27 × 103 km3 (Table 6), calculated using vertical contacts between vol- canoes yield overestimates for the volume of the youngest edifi ce and underestimates for the TABLE 6. ESTIMATED VOLUMES OF KA‘ENA, WAI‘ANAE, AND KO‘OLAU VOLCANOES volume of older ones. A further complication Volumes Literature estimates (×103 km3) (×103 km3)* to volume estimates is the presence of land- Vertical† 10° 30° B&J R&E Ito slide debris in two different forms—slumps and Main volcano volumes debris avalanche deposits (Moore et al., 1989). Ka‘ena 14.0 19.4 15.5 14.8 Robinson and Eakins (2006) argued that the Wai‘anae 18.2 26.6 20.9 25.0 52.9 31.0 Ko‘olau 31.3 17.5 27.1 20.9 31.7 34.3 boundaries of slumps are likely to be close to the original volcano outlines and therefore assumed Slide volumes Ka‘ena slide 3.3 the base of slumps to be the top of a compacted, Wai‘anae slump (Wai‘anae) 14.9 10.5 13.5 (0.6) 5.8 500-m-thick sediment layer overlying the oce- Wai‘anae slump (Ka‘ena) 4.4 1.4 § anic crust. In contrast, the basal contact of the Nu‘uanu debris slide 3.0 2.4 2.7 Nu‘uanu debris avalanche deposit is assumed to Total volcano volumes Ka‘ena 17.3 27.1 20.2 14.8 be the top of the surrounding archipelagic apron. Wai‘anae 33.1 37.1 34.4 25.0 53.5 36.8 Robinson and Eakins (2006) estimated the Ko‘olau 34.3 20.5 30.1 20.9 34.1 37.0 volume of Wai‘anae Volcano (including debris O‘ahu (all) 84.7 84.7 84.7 45.9 87.6 88.6 in the Wai‘anae and Ka‘ena landslides) to be *Literature values from B&J (Bargar and Jackson, 1974), R&E (Robinson and Eakins, 2006), and Ito et al. 53.5 × 103 km3, making it the fourth largest vol- (2013). All literature values assume vertical contacts. †Calculated volumes assuming vertical contacts to top of the Pacific plate (see text for details). Note R&E cano in the young end of the Hawaiian chain. report a volume of 0.6 × 103 km3 for a small debris field north of Ka‘ena Ridge, which we include in the Ka‘ena Using the same assumption of a vertical contact slide deposit. between Ka‘ena and Wai‘anae, we calculate the §Debris slide volume estimated using maximum seafloor depth around slide contacts for the base.

Geological Society of America Bulletin, Month/Month 2014 21 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al. comparable to the volcano volumes of line lavas in the Hawaiian Islands (Macdonald trate more than 5 km deep into the Ko‘olau edi- Kaho‘olawe, Lāna‘i, and East Moloka‘i given and Katsura, 1964; Zbinden and Sinton, 1988). fi ce for this to be the case, and it seems unlikely by Robinson and Eakins (2006) and those esti- Van der Zander et al. (2010) showed that the that much material from such depths would be mated for Mauna Kea and Hualālai by Lip- unique high-Mg, hornblende-biotite rhyodacite excavated and transported by the avalanche. man and Calvert (2013). Ka‘ena is substan- lava composition is best explained by hydrous tially larger than Māhukona (6–13 × 103 km3), partial melting of amphibolite in the lower Secondary Volcanism and although volume estimates for that volcano are crust beneath Wai‘anae Volcano. However, the Unconfi rmed 1956 Eruptive Activity poorly constrained (see discussion in Lipman silicic magmas have higher 208Pb/204Pb at lower in the Kaua‘i Channel and Calvert, 2013). 206Pb/204Pb than basaltic lavas of Wai‘anae Vol- Wai‘anae is the most voluminous of the cano, requiring that the source rock for the par- Forms of Hawaiian secondary volcanism three O‘ahu volcanoes. Depending on the tial melting must be isotopically distinct from include the classic rejuvenated-stage sequences slopes assumed for contact relations, Ka‘ena that which makes up most of Wai‘anae Volcano. on Ni‘ihau, Kaua‘i, East O‘ahu, East Moloka‘i, could be slightly smaller or somewhat larger Van der Zander et al. (2010) suggested that this and West Maui, fl ow fi elds on the South and than Ko‘olau. Our estimate for the volume of source material might be an early, isotopically North Hawaiian Arches (Lipman et al., 1989; Ko‘olau is greatly reduced from that of Robin- different, preshield phase of Wai‘anae Volcano. Clague et al., 1990), and scattered periph- son and Eakins (2006), owing to the presence The isotopic composition of the silicic samples eral alkalic provinces around the islands of of underlying Wai‘anae Volcano, as shown in and several less-evolved basalts from the earli- Ni‘ihau, Kaua‘i, and O‘ahu (Takahashi et al., the reconstruction of Moore and Clague (2002). est part of Wai‘anae lie almost exactly on the 2001; Hanyu et al., 2005; Dixon et al., 2008). Our total volume for O‘ahu agrees within uncer- Ka‘ena trend in 208Pb/204Pb versus 206Pb/204Pb Recognition of a new lava fi eld on the fl anks of tainty of bounding limits with that of Robinson (Fig. 9B), although the silicic lavas (Ws in Ka‘ena Ridge expands the known distribution and Eakins (2006). Fig. 9) have somewhat lower 87Sr/86Sr and higher and compositional range of young lava fl ows in 206 204 εNd than Ka‘ena samples at the same Pb/ Pb. the Kaua‘i Channel, between O‘ahu and Kaua‘i Implication for the Structural Evolution Our analysis of the overlapping relations among (Fig. 3). The Kaua‘i Channel lava fl ows share of Wai‘anae Volcano O‘ahu volcanoes indicates that Ka‘ena is likely compositional similarities with those in other to be present in the deep crust beneath the cen- secondary volcanic fi elds, including a range The recognition that Wai‘anae sits largely tral part of Wai‘anae (Fig. 14), especially if the of silica saturation from transitional basalt to on the fl ank of an earlier Ka‘ena Volcano helps slope of the contact is closer to 10° than 30°. nephelinite, generally depleted Sr-isotope ratios explain differences in rift zone development in We interpret the secondary trend within the at high 143Nd/144Nd, and extent of differentiation; different parts of Wai‘anae. Fiske and Jackson Wai‘anae isotopic arrays, culminating with the the Kaua‘i Channel lava fl ows reported here (1972) emphasized the effect of preexisting, but- silicic lava compositions (Fig. 9), to refl ect vari- range from 6 to 13 wt% MgO. tressing volcanoes on the geometry of Hawaiian able assimilation and partial melting of Ka‘ena Hawaiian secondary volcanic fi elds can form rift zones, and Lipman et al. (2006) considered crust in the evolution of the Wai‘anae magmatic over long periods of time. The longest-lived the geometric consequences of overlapping vol- system. All Wai‘anae lavas younger than 3.3 Ma examples span >2 m.y. on Kaua‘i and Ni‘ihau, canoes in the evolution of Kīlauea’s rift zone cluster within the main Wai‘anae isotopic fi eld, and ~800 k.y. on O‘ahu (Sherrod et al., 2007). structure. Zbinden and Sinton (1988) demon- indicating that interaction in the deep crust was Even the least-voluminous rejuvenated-stage strated the parallelism of dikes uniformly ori- limited to very early evolution, when the mag- sequence, the Lahaina volcanics of West Maui, ented N60°W in the NW quadrant of Wai‘anae matic conduits were possibly not well estab- was erupted over a period of ~300 k.y. (Tagami Volcano, in contrast to the much more radial lished, and during the short-term perturbation et al., 2003). Available data on the Kaua‘i Chan- disposition of dikes in the central and south- at 3.3 Ma that led to silicic volcanism (van der nel lava fl ows suggest that some of them were ern parts (Fig. 3). This fi nding remained enig- Zander et al., 2010). Later magmatic processes erupted as recently as 0.3–0.4 Ma. Given the matic in the context of Wai‘anae as an isolated apparently did not interact with the lower crust, generally long-lived nature of many Hawaiian (unbuttressed) Hawaiian volcano (e.g., Fiske and including the postshield stage of Wai‘anae with secondary volcanic provinces, and the relatively Jackson, 1972). We propose that Ka‘ena Volcano its expected deep magma chambers (Presley young ages for some Kaua‘i Channel lava fl ows, buttressed the northwestern part of Wai‘anae et al., 1997). the region has potential for recent volcanism. Volcano and infl uenced the orientation of its On 22 May 1956, the crew of a U.S. Navy northwest rift zone (Fig. 14). This infl uence is Implications for the Composition of plane observed a disturbance in the ocean not apparent in the southern part of Wai‘anae Ko‘olau Volcano and Nu‘uanu Debris ~60 km WNW of Ka‘ena Point on O‘ahu. Volcano, where dike orientations range from Avalanche Deposit This report triggered a frenzy of observations, SW to SSE (Stearns and Vaksvik, 1935; Stearns, both military and private, with contradictory 1939; Zbinden and Sinton, 1988) (Fig. 3). Our reconstruction of the extent of Wai‘anae accounts that were widely reported in the local beneath the Ko‘olau Volcano generally agrees newspapers. The only consistent observations Implications for the Geochemical Evolution with that of Moore and Clague (2002), who were of an area of ocean colored brownish-yel- of Wai‘anae Volcano showed that the Ko‘olau magma system was low or yellow-green. Some observers reported located within the fl anks of Wai‘anae Volcano. “fragments 5 to 10 cm in diameter” fl oating in The presence of Ka‘ena Volcano beneath According to our projections, the breakaway the water (Macdonald, 1959, p. 67). In addi- Wai‘anae also helps explain some enigmatic scarp for the Nu‘uanu landslide is underlain by tion, two (or possibly three) dead whales were features of Wai‘anae’s geochemical evolu- the fl anks of Wai‘anae Volcano, raising the ques- observed fl oating on the surface, as confi rmed tion. The Mauna Kūwale Rhyodacite and other tion of whether some of the debris in the Nu‘uanu by the captain of a fi shing vessel, who stated silicic lavas and dikes in the caldera region of landslide deposit might have been derived from that the bodies of the animals appeared to be Wai‘anae Volcano are the most-evolved sub alka- Wai‘anae Volcano. Faults would have to pene- undamaged, as if they were poisoned.

22 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano

A

22° 00′N A

21° 30′N

A′

20 km 21° 00′N 159°W 158°W 157°W A A′ 1000 B V.E. = 3:1 –1000 subaerial Ko‘olau submarine Wai‘anae subaerial –3000 ? arine Ka‘ena subm

Depth (m) –5000

30° 10° –7000 Top of oceanic crust 0 20 40 60 80 100 120 140 160 Distance along profile (km)

Figure 14. (A) Map of the O‘ahu region, showing schematic outlines and structural lineaments of Ka‘ena (blue), Wai‘anae (black), and Ko‘olau (green) Volcanoes at the base of the Hawaiian crust. Also shown are the distribution of major landslide deposits and their headwall scarps. Rift zones (color-coded by volcano) are shown as bold dashed lines; schematic caldera regions for Wai‘anae and Ko‘olau are shown as thin solid lines, and locations of weak gravity anomalies on Ka‘ena Ridge are shown as thin dashed lines. (B) Cross section along line A-A′ (lower panel) shows O‘ahu as an assemblage of overlapping volcanoes. Top of the oceanic crust is from Watts and ten Brink (1989) with an additional 500-m-thick sediment layer. Short-dashed lines depict 10° and 30° slopes of the Wai‘anae-Ka‘ena contact (see text for discussion). Interfi ngering contact relations between Ka‘ena and Wai‘anae are omitted. V.E.—vertical exaggeration.

Two days later, the discolored water was seen (Honolulu Star Bulletin, 1956). Macdonald ous origin,” he concluded that the evidence from the air to be rapidly dissipating. Geologists (1959) described these samples as “dark brown “appears to me to indicate a submarine erup- Gordon Macdonald and Agatin Abbott visited to black basaltic pumice containing phenocrysts tion” (p. 66, 68). the area aboard a small boat and found “no of olivine and labradorite” and reported that the Since 1956, the region between Kaua‘i and physical trace whatever” of the reported erup- fragments were “fresh and uneroded,” and that O‘ahu has been much better mapped, and sam- tion (Honolulu Advertiser, 1956a). A few days tests indicated that they would fl oat for only a ples from secondary lava fl ows in the Kaua‘i later, pieces of pumice were recovered from few hours (p. 68). Although Macdonald (1959) Channel have been recovered. The imprecise two locations on the windward side of O‘ahu described the disturbance as being of “dubi- location information for the 1956 “disturbance”

Geological Society of America Bulletin, Month/Month 2014 23 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al. suggests it occurred on the edge of Ka‘ena More than a week later, U.S. Air Force planes plexes on Ka‘ena Ridge indicate broadly dis- Ridge, where water depths increase rapidly from reported seeing yellow-green material that seminated volcanic activity. Ages of submarine ~1000 m to >3000 m (Fig. 1). Eruption in deep appeared to come from beneath the ocean sur- lithologies range from 4.9 to 3.6 Ma, consistent water is unlikely to produce pumice or disrup- face, but in this instance 215 miles northeast of with the thickness of Mn-Fe rinds. The end of tion of the ocean surface. All known secondary Honolulu (Honolulu Advertiser, 1956c; Hono- Ka‘ena subaerial activity is not well defi ned by volcanic features in the region are deep-water lulu Star Bulletin, 1956). existing data. pillow eruptions, including the newly discovered A volcanic eruption in the Kaua‘i Channel in The submarine to subaerial transition occurs pillow mounds, at water depths >2400 m (Fig. 1). 1956 remains unconfi rmed. Unusual ocean dis- more than 500 m deeper on the south side of Alternatively, an eruption on the shallow part of coloration pervaded a region of several hundred Ka‘ena Ridge than on the northwest, which sug- Ka‘ena Ridge might have produced pumice that kilometers around O‘ahu over a 2 wk period in gests differential subsidence or tilting 1.5°–2° to subsequently migrated with currents to the south late May 1956 but was likely related to unusual the south. This is the fi rst documented tilting of and west, as suggested by accounts of the time. biological activity at the time; whether or not an ancient, constructional horizon in a Hawaiian There are no identifi able young volcanic vents in such activity accounts for the deaths of whales volcano, but its orientation, roughly orthogonal the 10- to 20-m-resolution bathymetry or in the is unknown. Pumice recovered from O‘ahu to the strike of the Hawaiian Ridge, is enigmatic. limited refl ectivity data currently available for at the time is unlikely to have migrated from It contrasts with the orientation of modern reef the shallow part of Ka‘ena Ridge. the west, and it differed mineralogically from terraces, which mainly are tilted parallel to the Macdonald (1959) argued that any other known secondary lava samples from the Kaua‘i Hawaiian Ridge, in the direction of younger source for the “fresh” pumice found on O‘ahu Channel. If an eruption occurred in the shallow loading volcanoes. Whether or not differential was “diffi cult to visualize.” However, as noted by part of Ka‘ena Ridge in the past 60 yr, it is not subsidence of Ka‘ena Ridge is related to the Macdonald (1959), currents do not favor migra- apparent in the 10–20-m-resolution bathymetric repeated mass-wasting events that have affected tion of pumice from west to east in this area. data presently available. The identifi cation of it is unknown. Major oceanographic surface currents around the volumetrically smaller eruptive products will Taking into account overlapping relations Hawaiian Islands primarily fl ow from east and require higher-resolution mapping than is pres- between adjacent volcanoes, we estimate the northeast to the west. Furthermore, the olivine- ently available. volume of Ka‘ena Volcano to be in the range and plagioclase-bearing pumice fragments 20–27 × 103 km3, comparable to estimates for described by Macdonald (1959) do not match CONCLUSIONS Ko‘olau, East Moloka‘i, Lāna‘i, Kaho‘olawe, the mineralogy of known secondary lavas in the and Hualālai. The recognition that Wai‘anae Kaua‘i Channel, all of which contain olivine ± New visual observations and geochemical Volcano was built on the fl anks of an earlier clinopyroxene and are devoid of plagioclase. studies of samples collected from Ka‘ena and neighboring volcano explains the strongly ori- Pumice from locations outside the Hawai- Wai‘alu Ridges indicate a previously undocu- ented Wai‘anae NW rift zone, in contrast to the ian Islands occasionally washes up on Hawai- mented precursor volcano to the island of O‘ahu. more radial distribution of dikes in the southern, ian beaches. Most notable is the widespread Ka‘ena initiated as an isolated volcano in the relatively unbuttressed part of Wai‘anae. Pre- arrival of pumice from the 1952 Volcán Bárcena deep ocean and had an unusually prolonged his- viously enigmatic radiogenic isotope ratios in eruption at Isla San Benedicto, Revillagigedo tory of submarine growth before emerging in its highly silicic Wai‘anae lavas and dikes can now Islands, Mexico, which began arriving in rafts in late-shield stage of evolution. Ka‘ena Volcano be ascribed to partial melting and assimilation the early 1950s by westward circulation on the likely attained a maximum elevation ~1000 m of Ka‘ena material in the lower crust beneath North Equatorial Current (Jokiel, 1984; Jokiel above sea level ca. 3.5 Ma. Wai‘anae. and Cox, 2003). However, the San Benedicto Ka‘ena petrological evolution is dominated A reassessment of compositions of Ka‘ena peralkaline rhyolite pumice is greenish brown by an early phase of moderately high-SiO2, and Wai‘anae landslide deposits provides new in color and devoid of phenocrysts, unlike the low-FeO* tholeiitic eruptions. Close to the time constraints on their sources and ages of forma- basaltic pumice fragments described by Mac- of emergence, Ka‘ena evolved to include the tion. The Ka‘ena slide appears to be derived donald (1959). The source of the O‘ahu pum- eruption of transitional to alkalic lavas, with primarily from the upper reaches of Ka‘ena ice fragments therefore remains mysterious. We lower SiO2, and higher FeO*, P2O5, and other Volcano, including debris with late-shield com- have been unable to locate the original samples incompatible elements, and steeper chondrite- positions and low sulfur contents. We conclude in the collections of either the University of normalized REE patterns. Ka‘ena Volcano that much of the mass wasting occurred late in Hawai‘i or Hawaiian Volcano Observatory. likely reached a late-shield stage of volcanic the history of Ka‘ena, ca. 3.5 Ma. In contrast, Macdonald (1959) noted that discolored evolution, as indicated by interbedding between the compositions of limited samples from water in the Hawaiian area can arise from pro- the two distinctive chemical types and the lack the Wai‘anae slump suggest derivation from lifi c algal blooms and cited the failure of sensi- of an identifi able caldera. Ka‘ena and Wai‘alu Wai‘anae Volcano. However, headwall scarps tive seismographs on O‘ahu to detect any har- Ridge lavas form strongly correlated radio- around the Wai‘anae slump cross the boundary monic tremor or nearby earthquakes as evidence genic isotope arrays at relatively high values of between the Wai‘anae and Ka‘ena Volcanoes, that the disturbance might have been of biologi- 208Pb/204Pb over an extreme range of 206Pb/204Pb, requiring at least some contribution from Ka‘ena cal origin. Further support for this explanation in contrast to most lavas from neighboring to the Wai‘anae slump, which we estimate to be is found in additional newspaper reports about Wai‘anae Volcano. Samples from Ka‘ena and in the range 10–30 vol%. Ka‘ena material might the same time. On 25 May 1956, two separate Wai‘alu Ridges are chemically indistinguish- have been delivered to the offshore region dur- fi shing boats described greenish yellow material able, supporting the interpretation that these two ing an early phase of slumping and then buried in the ocean, unlike any that the crews had seen ridges represent different lineaments within a by later slumping along deeper scarps affecting before, variably said to be 3.5 h out and 15 miles single volcano. only Wai‘anae Volcano. off of windward O‘ahu, although the direction Broad, low-amplitude gravity anomalies, We discovered a new fi eld of basanite to was not disclosed (Honolulu Advertiser, 1956b). multiple topographic lineaments, and dike com- alkali basalt pillow mounds and lava fl ows that

24 Geological Society of America Bulletin, Month/Month 2014 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Ka‘ena Volcano mantle the south side of Ka‘ena Ridge, which Clague, D.A., 1987, Hawaiian xenolith populations, magma Society: Physical Sciences and Engineering, v. 342, supply rates and development of magma chambers: p. 121–136. extends the known distribution of secondary Bulletin of Volcanology, v. 49, p. 577–587, doi: 10 Frey, F.A., Garcia, M.O., and Roden, M.F., 1994, Geochemi- volcanism in the Kaua‘i Channel. Previous .1007 /BF01079963 . cal characteristics of Koolau volcano: Implications of radiometric dating indicated that some of this Clague, D.A., and Calvert, A.T., 2009, Postshield stage tran- intershield geochemical differences among Hawaiian sitional volcanism on Mahukona volcano, Hawai‘i: volcanoes: Geochimica et Cosmochimica Acta, v. 58, volcanism is as young as 0.34 Ma. These fi nd- Bulletin of Volcanology, v. 71, p. 533–539, doi: 10 .1007 p. 1441–1462, doi: 10 .1016 /0016 -7037 (94)90548 -7 . ings led us to reassess whether or not secondary /s00445-008 -0240 -z . Frey, F.A., Clague, D., Mahoney, J., and Sinton, J., 2000, volcanism could account for a disturbance of the Clague, D.A., and Moore, J.G., 1991, Geology and petrol- Volcanism at the edge of the Hawaiian plume: Petro- ogy of Mahukona volcano, Hawai‘i: Bulletin of Vol- genesis of submarine lavas from the North Arch vol- sea surface in 1956, which has been speculated canology, v. 53, p. 159–172, doi: 10 .1007 /BF00301227 . canic fi eld: Journal of Petrology, v. 41, p. 667–691, to be the result of a volcanic eruption in this Clague, D.A., and Sherrod, D.R., 2014, Growth and degrada- doi: 10 .1093 /petrology /41 .5 .667 . tion of Hawaiian volcanoes, in Poland, M.P, ed., Char- Gaffney, A., Nelson, B.K., and Blichert-Toft, J., 2004, Geo- region. Surface-water discoloration could have acteristics of Hawaiian Volcanoes: U.S. Geological chemical constraints on the role of oceanic lithosphere been of biological origin, and pumice collected Survey Professional Paper 1801, Chapter 3 (in press). in intra-volcano heterogeneity at West Maui, Hawaii: on O‘ahu beaches is petrographically different Clague, D.A., Moore, J.G., Torresan, M., Holcomb, R.T., Journal of Petrology, v. 45, p. 1663–1687, doi: 10 .1093 and Lipman, P.W., 1988, Shipboard Report for Hawaii /petrology /egh029 . from that of known secondary volcanics in the GLORIA Ground-Truth Cruise F2–88-HW, 25 Febru- Galer, S.J.G., 1999, Optimal double and triple spiking for Kaua‘i Channel, and it is unlikely to have been ary–9 March, 1988: U.S. Geological Survey Open-File high precision lead isotopic measurement: Chemical transported to O‘ahu against the prevailing cur- Report 88–292, 54 p. Geology, v. 157, p. 255–274, doi: 10 .1016 /S0009 -2541 Clague, D.A., Holcomb, R.T., Sinton, J.M., Detrick, R.S., (98)00203 -4 . rents. Moderately high-resolution bathymetric and Torreson, M.E., 1990, Pliocene and Pleistocene Galer, S.J.G., and O’Nions, R.K., 1985, Residence time of and refl ectivity data for the region of interest do alkalic fl ood basalts on the seafl oor north of the Hawai- thorium, uranium and lead in the mantle with impli- ian Islands: Earth and Planetary Science Letters, v. 98, cations for mantle convection: Nature, v. 316, p. 778– not reveal the presence of any large vents or lava p. 175–191, doi: 10 .1016 /0012 -821X (90)90058 -6 . 782, doi: 10 .1038 /316778a0 . fl ow fi elds that could have been emplaced dur- Coombs, M.L., Clague, D.A., Moore, G.F., and Cousens, Garcia, M.O., Muenow, D.W., and Aggrey, K.E., 1989, ing shallow-water, secondary volcanic eruptions B.L., 2004, Growth and collapse of Wai‘anae Volcano, Major element, volatile, and stable isotope geochemis- Hawai‘i, as revealed by exploration of its submarine try of Hawaiian submarine tholeiitic glasses: Journal of on Ka‘ena Ridge in the past century. fl anks: Geochemistry Geophysics Geosystems, v. 5, Geophysical Research, v. 94, p. 10,525–10,538, doi: 10 Q08006, doi: 10 .1029 /2004GC000717 . .1029 /JB094iB08p10525 . ACKNOWLEDGMENTS DePaolo, D.J., and Stolper, E.M., 1996, Models of Hawai- Garcia, M.O., Swinnard, L., Weis, D., Greene, A., Tagami, ian volcano growth and plume structure: Implications T., Sano, H., and Gandy, C., 2010, Petrology, geochem- Fred Duennebier was co-chief scientist on cruise of results from the Hawai‘i Scientifi c Drilling Project: istry and geochronology of Kaua‘i lavas over 4.5 Myr: KM06-03 of the R/V Kilo Moana, which was funded Journal of Geophysical Research, v. 101, p. 11,643– Implications for the origin of rejuvenated volcanism by the state of Hawai‘i in support of undergraduate 11,654, doi: 10 .1029 /96JB00070 . and the evolution of the Hawaiian plume: Journal of classes taught at the University of Hawai‘i. The student Dixon, J., Clague, D.A., and Stolper, E.M., 1991, Degassing Petrology, v. 51, p. 1507–1540, doi:10 .1093 /petrology history of water, sulfur, and carbon in submarine lavas /egq027 . participants on this cruise are thanked for their efforts at from Kilauea volcano, Hawaii: The Journal of Geol- Greene, A.R., Garcia, M.O., Weiss, D., Ito, G., Kuga, M., sea with station operations, data logging, and describ- ogy, v. 99, p. 371–394, doi: 10 .1086 /629501 . Robinson, J., and Yamasaki, S., 2010, Low-productivity ing samples. Fred also was chief scientist on R/V Kilo Dixon, J., Clague, D.A., Cousens, B., Monsalve, M.L., and Hawaiian volcanism between Kauai and Oahu: Geo- Moana cruise KM06-31 in 2006 and R/V Thompson Uhl, J., 2008, Carbonatite and silicate melt metasoma- chemistry Geophysics Geosystems, v. 11, Q0AC08, cruise TN 226 in 2008, which provided important sam- tism of the mantle surrounding the Hawaiian plume: doi:10 .1029 /2010GC003233 . ples for this project. Bruce Howe was chief scientist of Evidence from volatiles, trace elements and radiogenic Guillou, H., Carracedo, J.C., and Day, S., 1998, Dating of cruise KM11-16 and contributed to the success of that isotopes in rejuvenated-stage lavas of Niihau, Hawaii: the upper Pleistocene–Holocene volcanic activity of La expedition. The captains and crews of all these cruises Geochemistry Geophysics Geosystems, Q09005, doi: Palma using the unspiked K-Ar technique: Journal of 10 .1029 /2008GC002076 . Volcanology and Geothermal Research, v. 86, p. 137– are thanked for their efforts, especially the Jason team Doell, R.R., and Dalrymple, G.B., 1973, Potassium-argon 149, doi: 10 .1016 /S0377 -0273 (98)00074 -2 . from Woods Hole Oceanographic Institution, under ages and paleomagnetism of the Waianae and Koolau Guillou, H., Sinton, J., Laj, C., Kissel, C., and Széréméta, N., the leadership of Matt Heintz, who ensured the suc- Volcanic Series, Oahu, Hawaii: Geological Society of 2000, New K-Ar ages of the shield lavas from Waianae cessful completion of all remotely operated vehicle America Bulletin, v. 84, p. 1217–1242, doi: 10 .1130 volcano, Oahu, Hawaiian Archipelago: Journal of Vol- dives. Jesse Favia, Emilie Grau, Kristina Taylor, and /0016 -7606 (1973)84 <1217: PAAPOT>2 .0 .CO;2 . canology and Geothermal Research, v. 96, p. 229–242, Hannah Shelton helped with the processing of samples Fekiacova, Z., Abouchami, W., Galer, S.J.G., Garcia, M.O., doi: 10 .1016 /S0377 -0273 (99)00153 -5 . for chemical analysis. Ashton Flinders provided a grid and Hofmann, A.W., 2007, Origin and temporal evo- Hanyu, T., Clague, D.A., Kaneoka, I., Dunai, T.J., and of the gravity data used in Figure 10. Fred Duennebier, lution of Ko‘olau Volcano, Hawai‘i: Inferences from Davies , G.R., 2005, Noble gas systematics of sub marine isotope data on the Ko‘olau Scientifi c Drilling Project alkalic lavas near the Hawaiian hotspot: Chemical Scott Rowland, and Jim Kauahikaua encouraged a re- (KSDP), the Honolulu volcanics and ODP Site 843: Geology, v. 214, p. 135–155, doi:10 .1016 /j .chemgeo assessment of the 1956 submarine disturbance. Formal Earth and Planetary Science Letters, v. 261, p. 65–83, .2004 .08 .051 . reviews by Michelle Coombs, Dennis Geist, Robin doi: 10 .1016 /j .epsl .2007 .06 .005 . Haskins, E.H., and Garcia, M.O., 2004, Scientifi c drilling Holcomb, Dave Sherrod, and Asso ciate Editor Shan de Fiske, R.S., and Jackson, E.D., 1972, Orientation and growth reveals geochemical heterogeneity within the Ko‘olau Silva, and informal reviews by Alice Colman and Pete of Hawaiian volcanic rifts: The effect of regional struc- shield, Hawai‘i: Contributions to Mineralogy and Lipman led to substantial improvement of this paper . ture and gravitational stress: Proceedings of the Royal Petrology, v. 147, p. 162–188, doi:10 .1007 /s00410 -003 This research was supported by National Science Foun- Society of London, v. 329, p. 299–326, doi: 10 .1098 -0546 -y . dation grants EAR09-11301 and OCE 10-31485. This /rspa .1972 .0115 . Holcomb, R.T., and Robinson, J.E., 2004, Maps of Hawai- Flinders, A.F., Ito, G., and Garcia, M.O., 2010, Gravity ian Islands Exclusive Economic Zone Interpreted from is School of Ocean and Earth Science and Technology anomalies of the Northern Hawaiian Islands: Impli- GLORIA Sidescan-Sonar Imagery: U.S. Geologi- (SOEST) contribution 9014. cations on the shield evolutions of Kauai and Niihau: cal Survey Scientifi c Investigations Map 2824, scale Journal of Geophysical Research, v. 115, B08412, doi: 1:2,000,000, (http:// pubs .usgs .gov /sim /2004 /2824/). REFERENCES CITED 10 .1029 /2009JB006877 . Honolulu Advertiser, 1956a, Eruption eludes Macdonald, Flinders, A.F., Ito, G., Garcia, M.O., Sinton, J.M., Kaua- scientist fails to fi nd undersea volcano proof: Honolulu Abouchami, W., Hofmann, A.W., Galer, S.J.G., Frey, F.A., hikaua, J., and Taylor, B., 2013, Intrusive dike com- Advertiser, 25 May 1956, p. 1. Eisele, J., and Feigenson, M., 2005, Lead isotopes re- plexes, cumulate cores and exogenous growth of Honolulu Advertiser, 1956b, Two sampan crews see possible veal bilateral asymmetry and vertical continuity in the Hawaiian volcanoes: Geophysical Research Letters, volcanic residue: Honolulu Advertiser, 25 May 1956, Hawaiian mantle plume: Nature, v. 434, p. 851–856, v. 40, p. 3367–3373, doi: 10 .1002 /grl .50633 . p. A6. doi: 10 .1038 /nature03402 . Fodor, R.V., Frey, F.A., Bauer, G.R., and Clague, D.A., 1992, Honolulu Advertiser, 1956c, New underwater ‘eruption’ is Bargar, K.E., and Jackson, E.D., 1974, Calculated volumes Ages, rare-earth element enrichment, and petro genesis spotted: Honolulu Advertiser, 31 May 1956, p. 1. of individual shield volcanoes along the Hawaiian– of tholeiitic and alkalic basalts from Kahoolawe, Honolulu Star-Bulletin, 1956, New ‘eruption’ sighted; lava Emperor Chain: U.S. Geological Survey Journal of Hawaii : Contributions to Mineralogy and Petrology, from another found: Honolulu Star Bulletin, 31 May Research, v. 2, p. 545–550. v. 110, p. 442–462, doi: 10 .1007 /BF00344080 . 1956, p. 1. Charbit, S., Guillou, H., and Turpin, L., 1998, Cross cali- Frey, F.A., and Rhodes, J.M., 1993, Intershield geochemical Ito, G., Garcia, M.O., Smith, J.R., Taylor, B., Flinders, bration of K-Ar standard minerals using an unspiked differences among Hawaiian volcanoes: Implications A., Jicha, B., Yamasaki, S., Weiss, D., Swinnard, Ar measurement technique: Chemical Geology, v. 150, for source compositions, melting processes and magma L., and Blay, C., 2013, A low-relief shield volcano p. 147–159, doi: 10 .1016 /S0009 -2541 (98)00049 -7 . ascent paths: Philosophical Transactions of the Royal origin for the South Kauai Swell: Geochemistry Geo-

Geological Society of America Bulletin, Month/Month 2014 25 Geological Society of America Bulletin, published online on 2 May 2014 as doi:10.1130/B30936.1

Sinton et al.

physics Geosystems, v. 7, p. 2328–2348, doi: 10 .1002 Moore, J.G., and Clague, D.A., 2002, Mapping the Nuuanu Stearns, H.T., and Vaksvik, K.N., 1935, Geology and /ggge20159 . and Wailau landslides in Hawaii, in Takahashi, E., Groundwater Resources of the Island of Oahu, Hawaii: Jackson, M.C., Frey, F.A., Garcia, M.O., and Wilmoth, Lipman, P.W., Garcia, M.O., Naka, J., and Aramaki, Hawaii Division of Hydrography Bulletin 1, 113 p. R.A., 1999, Geology and geochemistry of basaltic lava S., eds., Hawaiian Volcanoes: Deep Underwater Per- Stille, P., Unruh, D.M., and Tatsumoto, M., 1986, Pb, Sr, Nd, fl ows and dikes from the Trans-Koolau tunnel, O‘ahu, spectives: American Geophysical Union Geophysical and Hf isotopic constraints on the origin of Hawaiian Hawaii : Bulletin of Volcanology, v. 60, p. 381–401, Monograph 128, p. 223–244. basalts and evidence for a unique mantle source: Geo- doi: 10 .1007 /s004450050239 . Moore, J.G., and Clague, D.A., 2004, Hawaiian submarine chimica et Cosmochimica Acta, v. 50, p. 2303–2319, Jokiel, P.L., 1984, Long distance dispersal of reef coral by manganese-iron oxide crusts: Geological Society of doi: 10 .1016 /0016 -7037 (86)90084 -0 . rafting: Coral Reefs, v. 3, p. 113–116, doi: 10 .1007 America Bulletin, v. 116, p. 337–347, doi:10 .1130 Strange, W.E., Woollard, G.P., and Rose, J.C., 1965, An /BF00263761 . /B25304 .1 . analysis of the gravity fi eld over the Hawaiian Islands Jokiel, P.L., and Cox, E.F., 2003, Drift pumice at Christmas Moore, J.G., Fornari, D.J., and Clague, D.A., 1985, Basalts in terms of crustal structure: Pacifi c Science, v. 19, Island and Hawai‘i: Evidence of oceanic dispersal from the 1877 Submarine Eruption of Mauna Loa, p. 381–389. patterns: Marine Geology, v. 202, p. 121–133, doi: 10 Hawaii : New Data on the Variation of Palagonitization Tagami, T., Nishimitsu, Y., and Sherrod, D.R., 2003, Re- .1016 /S0025 -3227 (03)00288 -3 . Rate with Temperatures: U.S. Geological Survey Bul- juvenated-stage volcanism after 0.6-m.y. quiescence Laj, C., Guillou, H., Szeremeta, N., and Coe, R., 1999, Geo- letin 1663, 11 p. at West Maui volcano, Hawai‘i: New evidence from magnetic paleosecular variation at Hawaii around 3 Moore, J.G., Clague, D.A., Holcomb, R.T., Lipman, P.W., K-Ar ages and chemistry of Lahaina volcanics: Jour- Ma from a sequence of 107 lava fl ows at Kaena Point Normark, W.R., and Torresan, M.E., 1989, Prodigious nal of Volcanology and Geothermal Research, v. 120, (O‘ahu): Earth and Planetary Science Letters, v. 170, submarine landslides on the Hawaiian Ridge: Journal p. 207–214, doi: 10 .1016 /S0377 -0273 (02)00385 -2 . p. 365–376, doi: 10 .1016 /S0012 -821X (99)00119 -3 . of Geophysical Research, v. 94, p. 17,465–17,484, doi: Takahashi, E., Moore, J.G., Yokose, H., Clague, D.A., Naka- Le Bas, M.J., Le Maitre, L.E., Streckeisen, A., and Zanettin, 10 .1029 /JB094iB12p17465 . gawa, M., Kani, T., Coombs, M., Moore, G., Harada, B., 1986, A chemical classifi cation of volcanic rocks Nakagawa, M., and Noguchi, N., 2003, Geochemistry of Y., Kunikiyo, T., and Robinson, J., 2001, A newly based on the total alkali–silica diagram: Journal of cluster of submarine lava cones, Kauai-Oahu Channel, recognized shield volcano southwest of Oahu Island, Petrol ogy, v. 27, p. 745–750, doi: 10 .1093 /petrology /27 Hawaii: Implications of magmatism at plume edge: Hawaii: Eos (Transactions, American Geophysical .3 .745 . Geochimica et Cosmochimica Acta, v. 67, no. 18, Union), v. 82, abstract V12B–0981. Lipman, P.W., and Calvert, A.T., 2011, Early growth of Supplement 322. Tanaka, R., Nakamura, E., and Takahashi, E., 2002, Geo- Kohala volcano and formation of long Hawaiian rift Presley, T.K., Sinton, J.M., and Pringle, M., 1997, Post- chemical evolution of Koolau volcano, Hawaii, in zones: Geology, v. 39, no. 7, p. 659–662, doi:10 .1130 shield volcanism and catastrophic mass wasting of the Takahashi, E., Lipman, P.W., Garcia, M.O., Naka, /G31929 .1 . Waianae volcano, Oahu, Hawaii: Bulletin of Volcanol- J., and Aramaki, S., eds., Hawaiian Volcanoes: Deep Lipman, P.W., and Calvert, A.T., 2013, Modeling volcano ogy, v. 58, p. 597–616, doi: 10 .1007 /s004450050165 . Under water Perspectives: American Geophysical growth on the island of Hawaii: Deep-water perspec- Ren, Z.-Y., Shibata, T., Yoshikawa, M., Johnson, K.T.M., and Union Geophysical Monograph 128, p. 311–332. tives: Geosphere, v. 9, no. 5, doi:10 .1130 /GES00935 .1 . Takahashi, E., 2006, Isotope compositions of submarine Torresan, M.E., Shor, A.N., Wilson, J.B., and Campbell, J., Lipman, P.W., Clague, D.A., Moore, J.G., and Holcomb, Hana Ridge lavas, Haleakala volcano, Hawaii: Implica- 1989, Cruise Report, Hawaiian GLORIA Leg 5, F5– R.T., 1989, South Arch volcanic fi eld—Newly iden- tions for source compositions, melting processes and 88-HW: U.S. Geological Survey Open-File Report tifi ed young lava fl ows on the sea fl oor south of the the structure of the Hawaiian plume: Journal of Petrol- 89–198, 56 p. Hawaiian Ridge: Geology, v. 17, p. 611–614, doi: 10 ogy, v. 47, p. 255–275, doi: 10 .1093 /petrology /egi074 . U.S. Navy, 1939, Bathymetric Chart of the North Pacifi c .1130 /0091 -7613 (1989)017 <0611: SAVFNI>2 .3 .CO;2 . Rhodes, J.M., and Vollinger, M.J., 2004, Composition of Ocean at 40° North Latitude, no. 5486, 500 Fathom Lipman, P.W., Sisson, T.W., Coombs, M.L., Calvert, A., and basaltic lavas sampled by phase-2 of the Hawaii Sci- Contour Interval: Washington, D.C., U.S. Navy Hydro- Kimura, J.-I., 2006, Piggyback tectonics: The long- entifi c Drilling Project: Geochemical stratigraphy and graphic Offi ce, scale 1:13, 600, 000. term growth of Kilauea on the fl ank of Mauna Loa: magma types: Geochemistry Geophysics Geosystems, van der Zander, I., Sinton, J.M., and Mahoney, J.J., 2010, Journal of Volcanology and Geothermal Research, v. 5, Q03G13, doi: 10 .1029 /2002GC000434 . Late shield-stage silicic magmatism at Wai‘anae Vol- v. 151, p. 73–108, doi: 10 .1016 /j .jvolgeores .2005 .07 Robinson, J.E., and Eakins, B.W., 2006, Calculated volumes cano: Evidence for hydrous crustal melting in Hawai- .032 . of Hawaiian shield volcanoes at the young end of the ian volcanoes: Journal of Petrology, v. 51, p. 671–701, Lonsdale, P., Dieu, J., and Natland, J., 1993, Posterosional Hawaiian Ridge: Journal of Volcanology and Geo- doi: 10 .1093 /petrology /egp094 . volcanism in the Cretaceous part of the Hawaiian thermal Research, v. 151, p. 309–317, doi:10 .1016 /j Walker, G.P.L., 1986, Koolau dike complex, Oahu: Intensity hotspot trail: Journal of Geophysical Research, v. 98, .jvolgeores .2005 .07 .033 . and origin of a sheeted-dike complex high in a Hawai- p. 4081–4098, doi: 10 .1029 /92JB02452 . Sherrod, D.R., Sinton, J.M., Watkins, S.E., and Brunt, K.M., ian volcanic edifi ce: Geology, v. 14, p. 310–313, doi: 10 Macdonald, G.A., 1959, The activity of Hawaiian volcanoes 2007, Geologic Map of the State of Hawai‘i: U.S. Geo- .1130 /0091 -7613 (1986)14 <310: KDCOIA>2.0 .CO;2 . during the years 1951–1956: Bulletin Volcanologique, logical Survey Open-File Report 2007–1089, scales Watts, A.B., and ten Brink, U.S., 1989, Crustal structure, v. 22, p. 3–70, doi: 10 .1007 /BF02596579 . 1:100,000 and 1:250,000. fl exure, and subsidence history of the Hawaiian Islands: Macdonald, G.A., 1968, Composition and origin of Hawai- Sinton, J.M., 1986, Revision of Stratigraphic Nomenclature Journal of Geophysical Research, v. 94, p. 10,473– ian lavas, in Coats, R.R., Hay, R.L., and Anderson, of Waianae Volcano, Oahu, Hawaii: U.S. Geological 10,500, doi:10 .1029 /JB094iB08p10473 . C.A., eds., Studies in Volcanology, A Memoir in Honor Survey Bulletin 1775-A, p. 9–15. Weis, D., Garcia, M.O., Rhodes, J.M., Jellinek, M., and of Howell Williams: Geological Society of America Sinton, J.M., 2005, Geologic mapping, volcanic stages Scoates, J.S., 2011, Role of the deep mantle in gen- Memoir 116, p. 477–522. and magmatic processes in Hawaiian volcanoes: Eos erating the compositional asymmetry of the Hawaiian Macdonald, G.A., and Katsura, T., 1964, Chemical com- (Transactions, American Geophysical Union), v. 86, mantle plume: Nature Geoscience, v. 4, p. 831–838, position of Hawaiian lavas: Journal of Petrology, v. 5, no. 52, Fall meeting supplement, abstract V51A–1471. doi: 10 .1038 /ngeo1328 . p. 82–133, doi: 10 .1093 /petrology /5 .1 .82 . Sinton, J.M., Grönvold, K., and Sæmundsson, K., 2005, Wilde, P., Chase, T.E., Normark, W.R., Thomas, J.A., and Macdonald, G.A., Abbott, A.T., and Peterson, F.L., 1983, Postglacial eruptive history of the Western volcanic Young, J.D., 1980, Oceanographic Data off Southern Volcanoes in the Sea (2nd ed.): Honolulu, Hawaii, Uni- zone, Iceland: Geochemistry Geophysics Geosystems, Hawaiian Islands: Lawrence Berkeley Laboratory Pub- versity of Hawaii Press, 517 p. v. 6, Q12009, doi: 10 .1029 /2005GC001021 . lication 359, scale 1:1, 030, 010 at 20°N Latitude, con- Mahoney, J.J., Nicollet, C., and Dupuy, C., 1991, Madagas- Smith, D.K., Kong, L.S.L., Johnson, K.T.M., and Reynolds, tour interval 100 meters, 1 sheet. car basalts: Tracking oceanic and continental sources: J.R., 2002, Volcanic morphology of the submarine Xu, G., Frey, F.A., Clague, D.A., Abouchami, W., Blichert- Earth and Planetary Science Letters, v. 104, p. 350– Puna Ridge, Kilauea Volcano, Hawaii, in Takahashi, Toft, J., Cousens, B., and Weisler, M., 2007, Geo- 363, doi:10 .1016 /0012 -821X (91)90215 -4 . E., Lipman, P.W., Garcia, M.O., Naka, J., and Aramaki, chemical characteristics of West Moloka‘i shield and Mahoney, J.J., Duncan, R.A., Tejada, M.L.G., Sager, W.W., S., eds., Hawaiian Volcanoes: Deep Underwater Per- postshield-stage lavas: Constraints on Hawaiian plume and Bralower, T., 2005, Jurassic-Cretaceous bound- spectives: American Geophysical Union Geophysical models: Geochemistry Geophysics Geosystems, v. 8, ary age and mid-ocean-ridge-type mantle source for Monograph 128, p. 125–142. Q08G21, doi: 10 .1029 /2006GC001554 . Shatsky Rise: Geology, v. 33, no. 3, p. 185–188, doi: Smith, J.R., 2002, The Kaena Ridge submarine rift zone off Zbinden, E.A., and Sinton, J.M., 1988, Dikes and the petrol- 10 .1130 /G21378 .1 . Oahu, Hawaii: Eos (Transactions, American Geophysi- ogy of Waianae volcano, Oahu: Journal of Geophysi- Mark, R.K., and Moore, J.G., 1987, Slopes of the Hawai- cal Union), v. 83, no. 47, Fall meeting supplement, ab- cal Research, v. 93, p. 14,856–14,866, doi: 10 .1029 ian Ridge: U.S. Geological Survey Professional Paper stract T62A–1300. /JB093iB12p14856 . 1350, v. 1, p. 101–107. Stearns, H.T., 1939, Geologic Map and Guide of the Island Moore, J.G., and Campbell, J.F., 1987, Age of tilted reefs, of Oahu, Hawaii: Hawaii Division of Hydrography SCIENCE EDITOR: NANCY RIGGS Hawaii: Journal of Geophysical Research, v. 92, Bulletin 2, scale 1:62,500. ASSOCIATE EDITOR: SHANAKA DE SILVA p. 2641–2646, doi: 10 .1029 /JB092iB03p02641 . Stearns, H.T., 1946, Geology of the Hawaiian Islands: Ha- MANUSCRIPT RECEIVED 28 MAY 2013 Moore, J.G., and Clague, D.A., 1992, Volcano growth and waii Division of Hydrography Bulletin 8, 105 p. REVISED MANUSCRIPT RECEIVED 19 FEBRUARY 2014 evolution of the island of Hawaii: Geological Society Stearns, H.T., 1978, Road Guide to Points of Geologic Inter- MANUSCRIPT ACCEPTED 7 APRIL 2014 of America Bulletin, v. 104, p. 1471–1484, doi: 10 .1130 est in the Hawaiian Islands (2nd ed.): Palo Alto, Cali- /0016 -7606 (1992)104 <1471:VGAEOT>2 .3 .CO;2 . fornia, Pacifi c Books, 100 p. Printed in the USA

26 Geological Society of America Bulletin, Month/Month 2014