The proximal part of the giant submarine Wailau landslide, Molokai, Hawaii

The proximal part of the giant submarine Wailau landslide,Molokai, Hawaii

David A. Clague a;�, James G. Moore b

a Monterey Bay Aquarium Research Institute, 7700 Sandholdt Road, Moss Landing, CA 95039, USAb U.S. Geological Survey, 345 Middle¢eld Road, Menlo Park, CA 94025, USA

Received 5 February 2001; received in revised form 18 June 2001; accepted 18 June 2001

Abstract

The main break-in-slope on the northern submarine flank of Molokai at 31500 to 31250 m is a shoreline featurethat has been only modestly modified by the Wailau landslide. Submarine canyons above the break-in-slope, includingone meandering stream, were subaerially carved. Where such canyons cross the break-in-slope, plunge pools mayform by erosion from bedload sediment carried down the canyons. West Molokai Volcano continued infrequentvolcanic activity that formed a series of small coastal sea cliffs, now submerged, as the island subsided. Lavas exposedat the break-in-slope are subaerially erupted and emplaced tholeiitic shield lavas. Submarine rejuvenated-stagevolcanic cones formed after the landslide took place and following at least 400^500 m of subsidence after the mainbreak-in-slope had formed. The sea cliff on east Molokai is not the headwall of the landslide, nor did it form entirelyby erosion. It may mark the location of a listric fault similar to the Hilina faults on present-day Kilauea Volcano. TheWailau landslide occurred about 1.5 Ma and the Kalaupapa Peninsula most likely formed 3307 5 ka. Molokai ispresently stable relative to sea level and has subsided no more than 30 m in the last 330 ka. At their peak, West andEast Molokai stood 1.6 and 3 km above sea level. High rainfall causes high surface runoff and formation of canyons,and increases groundwater pressure that during dike intrusions may lead to flank failure. Active shield or postshieldvolcanism (with dikes injected along rift zones) and high rainfall appear to be two components needed to trigger thedeep-seated giant Hawaiian landslides. = 2002 Elsevier Science B.V. All rights reserved.

Keywords: landslide; rejuvenated volcanism; submarine canyon; plunge pool; Hawaii

1. Introduction

A major north-directed landslide about 40 kmwide on the north submarine £ank of EastMolokai Volcano was ¢rst identi¢ed by examina-

tion of low-resolution bathymetric maps (Moore,1964). It was not, however, until systematicGLORIA side-scan-sonar surveys (Somers et al.,1978) were made in 1986 by the U.S. GeologicSurvey that the extent and nature of the landslide,named the Wailau debris avalanche, was mapped(Moore et al., 1989, 1994a,b). The GLORIA sur-vey showed that the slide was more than 100 kmlong and consisted of a large debris ¢eld onthe deep ocean £oor. The debris ¢eld mergeswith an adjacent large ¢eld derived from the

0377-0273 / 02 / $ ^ see front matter = 2002 Elsevier Science B.V. All rights reserved.PII: S 0 3 7 7 - 0 2 7 3 ( 0 1 ) 0 0 2 6 1 - X

* Corresponding author. Tel. : +1-831-775-1781;Fax: +1-831-775-1620.

E-mail address: [email protected] (D.A. Clague).

VOLGEO 2371 14-5-02

Journal of Volcanology and Geothermal Research 113 (2002) 259^287

www.elsevier.com/locate/jvolgeores

Nuuanu landslide from nearby Koolau Volcanoon Oahu. Moore and Clague (2002) reviewedthe history of mapping this landslide and theparallel development of ideas about its emplace-ment.The Wailau landslide has an apparent amphi-

theater 40 km wide at its head and has been esti-mated to have occurred 1.4 7 0.2 Ma (Moore etal., 1989) and 1.5 Ma (Clague et al., 2002). Theages and mechanics of such huge, apparentlycatastrophic, landslides are still largely unknown.Likewise, identi¢cation of the locations of someof the classic components of landslides (Varnes,1978), such as the main scarp and head, are notreadily apparent, which hinders dynamic model-ling of the landslide processes and consequencesof the landslides, such as their tsunamigenic po-tential. Submarine landslides in general pose sig-ni¢cant hazards and have been the subject ofnumerous recent studies (e.g., Lee, 1989; Hamp-ton et al., 1996). In this paper, we examine theupper region of the Wailau slide in an attempt tobetter understand how such large failures takeplace and the subsequent processes that modifythe landslide structures and complicate their inter-pretation.In 1993, two dives with NOAA’s Hawaii

Undersea Research Laboratory submersible Pis-ces V explored the submarine north £ank ofEast Molokai Volcano. The main purpose ofthese dives was to investigate the submarinebreak-in-slope, a candidate for the main landslidescarp, and a rejuvenated-stage (and post-land-slide) lava shield that grew near the upper partof the landslide and formed the Kalaupapa Pen-insula. Our goal was to better understand thegeologic history of this region. In 1998, a newhigh-resolution swath bathymetric and acousticbackscatter survey was conducted using a hull-mounted 30-kHz Simrad EM300 system (MBARIMapping Team, 2000). This survey, coupled withdigital elevation data of the island, provides un-precedented detail of the upper portion of theWailau landslide. In 1998 and 1999, SeaBeambathymetric mapping of the more distal parts ofthe landslide was completed but is described else-where (Naka et al., 2000; Smith et al., 2002;Moore and Clague, 2002).

2. Geology of Molokai Island

The island of Molokai, located east-southeastof Oahu, west-northwest of Maui and north ofLanai, consists of two major shield volcanoes(Fig. 1). West Molokai Volcano rises only 421 mabove sea level and has not been deeply eroded.The subaerial volcano consists of a relatively thinseries of tholeiitic shield lava £ows crosscut by afew dikes. The tholeiitic lavas are capped by athin veneer of postshield-stage hawaiite and mu-gearite cones and £ows. The shield and postshieldlavas have not been subdivided and are collec-tively called the West Molokai Volcanics (Lan-genheim and Clague, 1987). Two rift zones, ori-ented northwest and south-southwest, apparentlyjoin east of the summit. No thick, £at-lying lavaindicative of a summit caldera complex has beenexhumed in the shallow erosional canyons. Eastof the summit, a series of normal faults, orientedabout 140‡, step down to the east (Stearns andMacdonald, 1947). KÂr dating of one of theyoungest shield lavas, a strongly di¡erentiatedtholeiitic basalt £ow containing only 3.9^4.2%MgO, 53.5% SiO2, and 1.1% K2O, yielded anage of 1.907 0.06 Ma (McDougall, 1964). KÂrdating of four of the overlying postshield-stagehawaiite and mugearite £ows yielded ages be-tween 1.727 0.08 and 1.817 0.08 Ma (Clague,1987a). It is not known how much time prior to1.9 Ma is represented in the exposed part of theshield.The younger East Molokai Volcano rises

1514 m above sea level and is deeply eroded onits north £ank. The subaerial volcano consists ofthin-bedded £ows of tholeiitic basalt overlain by athick sequence of intercalated tholeiitic, transi-tional, and alkalic basalt £ows (Stearns and Mac-donald, 1947; Beeson, 1976; Clague and Beeson,1980) that together are called the lower memberof the East Molokai Volcanics (Langenheim andClague, 1987). These lavas are capped by a thicksequence of cones and £ows of postshield-stagehawaiite, mugearite, and trachyte that comprisethe upper member of the East Molokai Volcanics.Two tholeiitic lavas from high in the lower mem-ber of the East Molokai Volcanics have concor-dant KÂr ages of 1.527 0.07 Ma (McDougall,

VOLGEO 2371 14-5-02

D.A. Clague, J.G. Moore / Journal of Volcanology and Geothermal Research 113 (2002) 259^287260

1964) and another sample has been dated at1.767 0.07 (Naughton et al., 1980), although thisage appears too old based on its location relativeto the other dated samples. Three postshield-stagealkalic lavas are dated at 1.35^1.49 Ma (McDou-gall, 1964). Thus, shield-building on West Molo-kai ended about 380 ka earlier than on East Mo-lokai and postshield volcanism on both volcanoescontinued for roughly 0.2 Ma after the end of

shield-building. The duration of shield activityprior to 1.52 Ma has not been determined.The northern coast of East Molokai is marked

by a high sea cli¡, the steepest slope in Hawaii(Mark and Moore, 1987), that is incised by aseries of deep canyons (the largest, from west toeast, are Waialeia, Waikolu, Pelekunu, and Wai-lau; Fig. 2) that mainly cut into a sequence ofponded lavas interpreted to represent a caldera

Fig. 1. Map of the Molokai and Oahu region showing major landslides relative to the islands. The map merges SeaBeam 2100bathymetry (Smith et al., 2002; Moore and Clague, 2002), Simrad EM300 bathymetry (MBARI Mapping Team, 2000; Dartnelland Gardner, 1999), and nearshore survey north of Molokai from Mathewson (1970). The general regions of blocks comprisingthe Nuuanu and Wailau landslides are labeled. Box shows the area enlarged in Fig. 4.

VOLGEO 2371 14-5-02

D.A. Clague, J.G. Moore / Journal of Volcanology and Geothermal Research 113 (2002) 259^287 261

complex (Stearns and Macdonald, 1947). Subse-quent mapping and paleomagnetic studies on thenorth coast and within the canyons (Holcomb,1985) suggest that the caldera complex maybe signi¢cantly larger than originally mapped(Stearns and Macdonald, 1947). Two dike swarmsmark the locations of rift zones on the west edgeof the caldera complex, oriented west-northwest,and another identi¢es a rift zone on its east edge,oriented east-northeast. The steep sea cli¡ rises asmuch as 1000 m between Pelekunu and Waikolucanyons. Several geologists inferred that the cli¡formed by erosion and that its great height indi-cated a long period of erosion (Dana, 1890, whoactually referred to the Nuuanu Pali on Oahu;Wentworth, 1927; Stearns and Macdonald,1947; Macdonald and Abbott, 1970). The heightof the cli¡ decreases dramatically both east andwest from this area. Near Cape Halawa at thenortheastern point of the island, the sea cli¡ isas little as 150 m high. Likewise, to the west the

sea cli¡ slowly decreases in height and is about100 m high due north of the airport and disap-pears completely about 2 km farther west. Southof the Kalaupapa Peninsula the cli¡ rises to anelevation of about 640 m, but the base of the cli¡is at an elevation of 75 m due to ponding of lavaagainst the cli¡ when the peninsula formed. Nearthe highest part of the cli¡, the average slope isabout 55‡. Behind the Kalaupapa Peninsula, thesea cli¡ is located about 500 m farther north thanon either side of the peninsula. Moore et al.(1989) identi¢ed the sea cli¡ as the main scarpof the Wailau landslide, modi¢ed somewhat bysouthward retreat caused by marine erosion.After the northern sea cli¡ had largely formed,

a rejuvenated-stage shield named Kalaupapa,123 m above sea level, erupted at its base toform the Kalaupapa Peninsula (Stearns and Mac-donald, 1947); the peninsula projects 4 km northof the cli¡ and is 5 km wide at its point of attach-ment (Fig. 3). The Kalaupapa shield is indented

Fig. 2. Bathymetry north of East Molokai (Shepard and Dill, 1966) showing the heads of many of the submarine canyons andthe general location of the break-in-slope. Locations referred to in the text are labeled. The locations of dike swarms on EastMolokai (Stearns and Macdonald, 1947) are indicated by heavy dashed double lines.

VOLGEO 2371 14-5-02


by a crater, named Kauhako Crater, about 0.5 kmin diameter that reaches below sea level and is¢lled with brackish water (Walker, 1990). Thiscrater fed the major lava £ows that constructedat least the subaerial part of the shield. An openlava channel extends from the north side of thecrater 1 km down the north £ank of the shieldand there transforms into a lava tube that fedtumuli and secondary vents that continue an addi-tional 1.3 km north to the shoreline. A blockylava £ow fed from the tube at about 30 m eleva-tion £owed about 1 km west to near the shoreline.A smaller secondary lava tube branches from theprimary one about 1 km north of the crater rimand can be followed north-northeast to near theeast shoreline of the peninsula; it is marked byelongate tumuli and eight collapsed skylights(Coombs et al., 1990). The entire KalaupapaPeninsula appears to be a single monogeneticvent (Coombs et al., 1990). The lava is alkalicbasalt and basanite in composition and has been

dated by KÂr at 0.357 0.03 to 0.577 0.02 Ma(Clague et al., 1982), although a monogenetic ori-gin suggests that the entire shield should haveformed in tens to hundreds of years, rather thanthe 220 ka suggested by the KÂr ages. A smallcone of ash named Pu’u ’Ula located 1.4 kmsouthwest of Kauhako Crater at the cli¡ base isinterpreted to be a separate rejuvenated-stage ventthat erupted only pyroclastic material (Walker,1990; Coombs et al., 1990).

3. Morphology of the north submarine slope ofMolokai from early surveys

Early surveys, including U.S. Navy Oceano-graphic O⁄ce chart BCO4N of the Hawaiian re-gion (Moore, 1964) and derivatives of it (Wilde etal., 1980), a chart of the North Molokai region toabout 2000 m depth by A. Malaho¡ and C.D.Hollister (written communication, 1987, published

Fig. 3. Topography and bathymetry of the Kalaupapa shield volcano showing the track of dive P5-252 on the north submarineslope. Contour interval is 100 m above and below sea level. Numbers along dive track are sample locations. Contours afterMathewson (1970). Location of image shown in Fig. 4A.

VOLGEO 2371 14-5-02


VOLGEO 2371 14-5-02


by Moore et al., 1989), revealed a prominent sub-marine scarp, concave to the north away from thevolcano (Shepard and Dill, 1966; Fig. 2). Thescarp begins at a depth of 1200^1500 m and de-scends to a depth of 2100^2500 m. Towards thesides of the horseshoe, the top of the scarp be-comes progressively deeper. Above the scarp theslope is 3.4^4.8‡, typical of the subaerial shields ofHawaiian volcanoes. The scarp has slopes of 17^20‡, and is 700^2200 m high. Below the scarp thebottom becomes less steep and smoother beforeencountering the irregular topography of thelandslide blocks.A series of 11 deep canyons extend from above

sea level, across this scarp, and down to at least2000 m depth (Shepard and Dill, 1966; Mathew-son, 1970). These £at-£oored canyons are 1^2 kmwide, up to about 200 m deep, and commonly arein line with major canyons above sea level on EastMolokai Volcano. The canyon heads are locateda few km o¡shore and the canyons extend about13 km to the edge of the terrace at roughly 31300m. The submarine canyons are best developedabove the break-in-slope; they become narrowerbelow it and generally disappear below about2000 m (Fig. 2). These major canyons, apparentlycut into volcanic bedrock, are similar in morphol-ogy to Hawaiian canyons on land. The upperparts, shallower than 31300 m, are regarded ashaving been carved subaerially when the volcanostood at least 1300 m higher than at present(Shepard and Dill, 1966; Mathewson, 1970; An-drews and Bainbridge, 1972; Coulbourn et al.,1974). The deeper parts, extending down to about32000 m, are regarded as having been carvedsubaqueously. Moore (1987) and Moore andCampbell (1987) extended this argument to sug-gest that the break-in-slope represents the shore-line at the time shield-building volcanic activityon East Molokai waned.A prominent terrace extends 4^5 km o¡shore

down to 31407 20 m on the north submarineslope of the island (Fig. 2). The 3140 m terrace

is apparently a combined wave-cut platform andreef-sediment bank, developed during low standsof the sea during Pleistocene glacial maxima. Thelower limit of the bench is deeper (V155 m) nearthe western end of the island compared to the eastend (V120 m) (Fig. 2).In addition to the shallow terrace, canyons and

steep scarp, the region a¡ected by the landslidewas modi¢ed near sea level by the growth of Ka-laupapa shield between 0.35 and 0.57 Ma (Clagueet al., 1982). The shield grew near the shorelineafter the large submarine canyons were cut andforms a peninsula attached to the island of Mo-lokai (Fig. 1). Below sea level the north £ank ofthe Kalaupapa shield forms a prominent bulgedown to about 500 m depth that has a volumeof about 3 km3 above the estimated pre-volcanoregional slope (Coombs et al., 1990), and has¢lled in the upper parts of two canyons (Fig. 3).There is a prominent break-in-slope at about 340m depth on the Kalaupapa shield, but the 3140m terrace is missing. The 340 m break-in-slopeapparently marks the position of sea level at thetime the volcano grew (Moore, 1987).

4. Morphology of the submarine north slope ofMolokai from Simrad EM300 data

The new high-resolution bathymetry and acous-tic backscatter data, shown in Fig. 4A,B, revealmany features not apparent in the lower-resolu-tion bathymetry. In particular, the depth to thebreak-in-slope and how it varies across the slidecan be determined more accurately and the shapeof the canyons is revealed in more detail. Many ofthese ¢ne-scale features demonstrate past posi-tions of sea level and are critical in determiningthe extent of the landslide. These features are de-scribed and illustrated in this section. The imagesthat follow are arranged from west to east andhave been selected to show speci¢c features.A meandering stream cuts into the terrace near

Fig. 4. (A) Bathymetry illuminated from the northwest with 500-m contours of north Molokai region based on Simrad EM300data. Other bathymetry is shown only as contours. Areas of Figs. 3 and 5^11 are shown and labeled. (B) Sidescan of the sameregion collected with the Simrad EM300 system. Lighter shades indicate greater acoustic re£ectance.

VOLGEO 2371 14-5-02


the west edge of the survey (Fig. 5). The sharpestloop in the canyon has nearly cut through into thecanyon to the east. The £oor of the canyon in themeander is about 50 m deeper than the walls andhas high backscatter, as do the £oors of all thecanyons, indicating harder material than the ter-

race into which the canyons are cut. We suspectthat the canyon £oors are either sand-covered orexposed bedrock, in contrast to the mud-coveredterrace. The canyon plunges 350 m down a steepcli¡ at the scarp with its rim at 1175 m depth andforms a 25-m-deep plunge pool at the base. The

Fig. 5. Bathymetry illuminated from the northwest with 25-m contours showing meandering stream canyon. Location of imageshown in Fig. 4A. Canyon cutting has caused scarp to recede about 0.5 km and has produced plunge pool at base of scarp.

VOLGEO 2371 14-5-02


break-in-slope varies from about 1125 m at thewestern edge of the image to about 1300 m atthe eastern edge. The eastern part of the terracehas several small steps, one at about 1000 m and adeeper one at about 1075 m. The deeper step maybe a carbonate reef formed at the shoreline,whereas the shallower one extends to the east,as shown in Fig. 6, and is probably a sea cli¡formed when a lava £ow entered the ocean.To the east of the meandering stream, the ter-

race is cut by the several small canyons and ischaracterized by numerous small contour-parallelsteps in bathymetry (Fig. 6). These steps probablyformed when lava entered the ocean and spread

out along the shoreline. The steps are thereforecoastal sea cli¡s formed after the main break-in-slope as volcanic activity waned at the end ofshield-building volcanic activity on West Molo-kai, slightly before 1.9 Ma. These shorelineshave subsequently subsided to their present depth.Two additional canyons cross the terrace, butcannot be traced to the present shoreline, norare there any subaerial canyons near their prox-imal ends. The main break-in-slope continues todeepen from west to east and is about 31250 mdeep at the eastern edge of the image. The shore-line terraces also deepen from west to east.The terrace northwest of the Kalaupapa Penin-

Fig. 6. Bathymetry illuminated from the northwest with 100-m contours showing small steps on the terrace formed when lava en-tered the ocean and built shoreline lava deltas. Location of image shown in Fig. 4A.

VOLGEO 2371 14-5-02


sula is shown in Fig. 7. Three major canyonscross the terrace in this region. The deepest cen-tral canyon is about 200 m deep and its £at £ooris about 700 m wide. This canyon plunges from1300 to 1650 m over a steep cli¡ at the edge of theterrace and forms a plunge pool at the cli¡ base(Fig. 8). The pool is about 600 m in diameter and

the bottom of the pool is about 90 m deeper thanthe sill. The canyons have cut back the mainbreak-in-slope scarp to form deep reentrantsabove the plunge pools (Fig. 7). The terraceedge becomes less distinct east of this canyonalthough the slope still has a steeper zone betweengentler slopes deeper than about 2000 m and shal-

Fig. 7. Bathymetry illuminated from the northwest with 100-m contours of the terrace northwest of the Kalaupapa Peninsulashowing three major canyons that cross the terrace in this region. Location of image shown in Fig. 4A. Canyon cutting hascaused the scarp to recede about 1 km and has produced a large plunge pool at base of scarp.

VOLGEO 2371 14-5-02


Fig. 8. Bathymetry illuminated from the northwest with 10-m contours showing 90-m-deep plunge pool at base of break-in-slopebelow submarine canyon. Location of image shown in Fig. 4A.

VOLGEO 2371 14-5-02


lower than 1300 m. The eastern canyon has trib-utaries that coalesce downslope. The western can-yon has a ridge along its eastern margin.The terrace east of Fig. 7 is shown in Fig. 9.

The region is dominated by two canyon systemswith the eastern one showing numerous strands,principally the lower reaches of Waikolu andPelekunu canyons, that coalesce to the north.The distal end of the £oor of this combined largecanyon is remarkably £at and nearly 1 km wide.It has cut down through the break-in-slope toproduce a more gradual slope instead of a steepcli¡ and plunge pool as seen to the west. Thewestern canyon also has a £at £oor with a smallV-shaped notch and has cut down throughthe break-in-slope. The break-in-slope is less dis-tinct than farther west and the canyons aremore numerous and larger in this region o¡shorefrom the higher, and rainier, East Molokai high-lands.Two complex volcanic cones are located on the

western side of lower Waikolu canyon (Fig. 9).The northeastern cone consists of a ring-likestructure with a small central peak, although theeastern part of the cone is truncated by the can-yon. The southwestern cone is steeper on the sidefacing the canyon as well. Both cones may havebeen modi¢ed by slumping as erosion in the can-yon undercut the cones. An area around anddownslope from the cones has high backscatter(seen in Fig. 4B and outlined in Fig. 9) andmost likely consists of the £ows erupted fromthe cones. No other comparable areas of highbackscatter or conical structures occur in the sur-vey region.The canyons in Fig. 9 feed onto a smooth sedi-

ment basin that slopes down to the north andthen to the east (Fig. 10). This sedimentary basinis ponded behind a 6-km block in the Wailaulandslide with the sediment apparently all fedfrom the terrace through the eastern canyonshown in Fig. 9, since a low ridge separates thisdrainage from that to the west. The 6-km blockhas a steep southern face and a more gentlenorthern face with numerous small radiating can-yons that further indicate the e⁄cacy of post-landslide submarine erosion. Hence, landslidingaccelerated erosion by the upper reaches of these

streams, but landslide blocks dammed the streamsfurther down. Eventually when the dammed basinwas ¢lled with sediment, the stream overtoppedthe eastern side of the block and sediment spilledout to the north. The break-in-slope is less sharpin this area than farther west and has been modi-¢ed by numerous post-landslide slumps that leftembayments, the largest being 1^2 km wide.The eastern part of the terrace north-northeast

of the eastern end of Molokai Island with abreak-in-slope at about 31500 m deep is crossedby several large, £at-£oored canyons that haveeroded the break-in-slope (Fig. 11). In addition,a less obvious change at about 3500 m is anothershoreline feature, most probably a drowned car-bonate reef on the northwest slope of Maui Is-land. Several other arcuate structures on the ter-race are probably relict shoreline features.

5. Direct submersible observations

5.1. The north submarine slope of Molokai

Pisces V dive P253 on October 10, 1993, began13 km north-northeast of the north tip of theKalaupapa Peninsula (Fig. 9). During the dive,the scarp below the break-in-slope was ascendedand the surface of the terrace was examined. Thesubmersible touched down on a steep slope lit-tered with angular rubble at a depth of 31780m (Fig. 9). From this point up to 31600 m mas-sive outcrops of lava alternate with slopes litteredwith angular talus. From 31600 m to about31400 m the sloping bottom is completely man-tled by partly indurated layers of carbonate-ce-mented volcanic sand; no lava outcrops wereseen. From 31400 m to the end of dive at31205 m pillow lava mounds partly covered bycarbonate sand occur.

5.2. The £ank of the Kalaupapa Peninsula

Pisces V dive P252 on October 8, 1993, toucheddown at 3625 m on the north £ank of the Ka-laupapa Peninsula, and moved upslope towardthe south (Fig. 3). The bottom is mantled with abu¡-colored mud, showing faint ripple marks.

VOLGEO 2371 14-5-02


Channels in this sediment expose indurated silt-stone with many fragments of rock up to 0.6 m insize that appear to be pillow fragments. At3550 m, a foreset-bedded pillow breccia was en-countered that extends up to 3475 m. No out-crops occur from 3475 m upward to 390 m. A

silt-covered sequence of terraces in this zone ismantled with white carbonate rubble that in-creases in size upwards.Bold outcrops of in-place pillow lava were en-

countered at 390 m. This outcrop was followedup to about 370 m where it is overlain by car-

Fig. 9. Bathymetry illuminated from the northwest with 100-m contours of the terrace east of Fig. 7. Location of image shownin Fig. 4A. The track of Pisces V dive P5-253 is shown with sample locations labelled. A pair of volcanic cones, the only ones inthe surveyed region, are labelled C. An area with high backscatter, interpreted to be lava £ows from the two volcanic cones, isenclosed by a heavy line.

VOLGEO 2371 14-5-02


Fig. 11. Bathymetry illuminated from the northwest with 100-m contours of eastern part of the terrace north-northeast of theeast end of Molokai Island. Location of image shown in Fig. 4A.

Fig. 10. Bathymetry illuminated from the northwest with 100-m contours of the nearest landslide block in the Wailau landslide.Location of image shown in Fig. 4A.

VOLGEO 2371 14-5-02


VOLGEO 2371 14-5-02


bonate rubble and sand. Above, other outcrops ofpillow lava were followed up to 350 m and ob-servations from the viewports suggest that theycontinue upwards to at least 345 m. Observationsended when the submersible left the bottom at350 m because strong surge related to surfacewaves made shallower work hazardous.

6. Petrography and geochemistry of submarinesamples

6.1. The north submarine slope of Molokai

Twelve samples were collected during diveP253. Whole-rock major- and trace-element anal-yses of nine samples are presented in Table 1A,B,microprobe analyses of glass rinds from ¢ve sam-ples are presented in Table 2, and microprobeanalyses of 15 glass inclusions in olivine are pre-sented in Table 3. The samples are divided intotwo groups, shown in Fig. 12, those from belowthe 31400 m scarp (31800 to 31615 m) andthose above the scarp (31350 to 31205 m). Thesamples from below the scarp are subaeriallyerupted tholeiitic basalt erupted during the shieldstage, and those above the scarp are subaqueouslyerupted alkalic basalt erupted during the rejuven-ated stage (Fig. 12A).The subaerial nature of lava from below the

terrace is supported by the lack of pillow struc-tures, the abundant vesicles in these samples, andthe low sulfur content of the one analyzed glass(Fig. 12B). Subaerially erupted lavas, like sampleP253-1 (Table 2), contain 6 250 ppm S whereaslavas erupted deeper than a few hundred metersgenerally contain s 800 ppm S (Moore andClague, 1987; Moore and Thomas, 1988). Thelow S contents (50, 60, and 220 ppm S, Table 3)of three glass inclusions from this sample suggestthat the olivine crystallized during subaerial stor-age in a summit lava pond or during emplacementof the £ow following subaerial degassing. The ma-jor element compositions of the four analyzedtholeiitic to transitional lavas are similar to thoseof lava £ows that comprise the main cli¡ sectionof East Molokai Volcano (Beeson, 1976; Clagueand Beeson, 1980; Clague et al., 1983). These

lavas are part of the late subaerial shield of thevolcano that erupted from the summit or thewest-northwest rift zone and £owed northward,towards the shoreline.The alkalic pillow lavas on the terrace contrast

markedly with the lava collected at greater depthon the same dive. These lavas are geochemicallysimilar to rejuvenated-stage lavas that comprisethe Kalaupapa Peninsula (Clague et al., 1982)and other rejuvenated-stage lavas in the islands(Fig. 13, and see summary in Clague, 1987b).The high sulfur content of the glass rinds andglass inclusions (Fig. 12B) indicates that theyerupted below sea level. These lavas erupted aftersubaerial cutting of the canyons, after formationof the 31400 m scarp that they mantle, and aftersubstantial subsidence had occurred, and presum-ably after the northern sea cli¡ was largelyformed. Their eruption was followed by collapseof some lava into the canyons and further suba-queous downcutting of the canyons. This is ourfavored model, but there are no unequivocal ob-servational data that demonstrate that the can-yons initially formed by subaerial erosional pro-cesses.

6.2. The £ank of the Kalaupapa Peninsula

Nine samples were collected during dive P252.Major- and trace-element analysis of one whole-rock sample (Table 1) and microprobe analyses ofglass rinds from four samples (Table 2) show thatthese samples are subaerially erupted alkalic ba-salt that £owed into the sea and was quenched.Microprobe analyses of 12 glass inclusions in oli-vine crystals are presented in Table 3.The whole-rock analyses of alkalic basalt sam-

ples collected on land (Clague et al., 1982) repre-sent material analyzed over a period of years bydi¡erent methods in four laboratories and, notsurprisingly, show some scatter (Fig. 12A). How-ever, the two samples collected within the KahukoCrater of the Kalaupapa shield are distinctive inhaving higher FeO, TiO2, K2O, P2O5, and lowerAl2O3 and SiO2 than the other subaerial rocks.They probably represent a somewhat di¡erentcomposition plastering the crater walls at a latestage in the eruption.

VOLGEO 2371 14-5-02


Fig. 12. (A) Alkali^silica plot of submarine lava collected from north of Molokai. The diagonal line separates tholeiitic and al-kalic basalt compositions. Whole rock, pillow rind glass, and glass inclusions in olivine are shown. (B) Sulfur as a function ofcollection depth for glass rinds and glass inclusions in olivine crystals collected north of Molokai.

VOLGEO 2371 14-5-02


Table 1AWhole-rock chemical analyses

Sample: P252-2 P253-2 P253-3 P253-4 P253-8 P253-9 P253-10A P253-10A P253-11A P253-11B P253-12Depth (m): 3560 31755 31680 31615 31450 31350 31280 (Repeat) 31245 31245 31205

SiO2 46.1 48.4 47.0 48.3 46.4 46.1 46.1 43.9 44.0 45.9Al2O3 12.5 13.2 14.3 13.7 15.5 14.9 15.5 11.5 11.5 14.8FeO 10.6 8.17 3.94 8.22 7.48 7.48 7.04 9.39 9.63 7.38Fe2O3 1.82 3.92 8.32 3.97 4.09 3.99 4.78 3.17 3.10 4.40MnO 0.18 0.18 0.17 0.19 0.17 0.15 0.25 0.17 0.17 0.15MgO 13.7 8.33 7.51 7.92 6.64 8.29 6.29 15.9 15.7 7.89CaO 9.84 10.7 11.4 10.9 12.4 12.0 12.5 9.50 9.48 12.0Na2O 2.35 2.19 2.43 2.11 2.94 2.90 3.01 2.27 2.30 2.92K2O 0.33 0.23 0.28 0.31 0.64 0.62 0.64 0.56 0.57 0.66TiO2 1.42 2.55 2.58 2.63 2.00 1.96 2.01 1.65 1.67 1.97P2O5 0.20 0.26 0.27 0.24 0.24 0.22 0.25 0.24 0.24 0.25CO2 0.040 ^ ^ ^ ^ ^ ^ ^ ^ ^H2O+ 0.43 0.46 0.47 0.67 0.51 0.49 0.70 0.70 0.67 0.69H2O3 0.20 0.58 0.43 0.29 0.25 0.23 0.26 0.17 0.20 0.22Total 99.7 99.9 99.8 100.1 99.8 100.0 100.0 99.8 99.9 99.9

X-ray £uorescence analysisNi 409 116 101 109 100 224 104 541 537 168Cr 594 352 288 350 241 343 237 546 543 287Sc 25 39 34 34 24 26 31 25 25 31V 247 311 312 318 337 318 337 277 264 328Ba 148 35 53 36 208 215 235 185 189 216Rb 5 0 2 1 9 10 10 9 6 10Sr 277 285 306 288 352 346 372 308 306 354Zr 70 135 142 138 100 98 102 84 86 98Y 19 28 31 29 24 22 22 17 19 22Nb 13.9 13.4 12.6 12.8 19.9 19.6 20.3 17.4 18.2 18.5Ga 20 19 21 21 20 22 23 20 18 23Cu 100 114 131 106 98 137 91 83 87 92Zn 94 109 110 107 96 s 102 94 97 101 92

ICP-MS analysesLa 11.22 10.53 10.66 10.09 14.15 14.16 14.84 14.55 12.86 13.17 14.54Ce 22.35 26.08 25.06 23.98 28.82 28.91 29.93 29.44 25.71 26.40 29.03Pr 2.76 3.65 3.59 3.54 3.65 3.61 3.76 3.64 3.21 3.26 3.65Nd 12.60 18.43 18.37 17.86 16.83 16.50 17.36 16.78 14.76 14.99 16.94Sm 3.49 5.69 5.74 5.56 4.68 4.62 4.86 4.72 3.92 4.07 4.69Eu 1.28 2.00 2.03 2.00 1.65 1.63 1.72 1.69 1.38 1.45 1.64Gd 3.80 6.15 6.29 6.05 5.00 4.85 4.95 5.10 4.18 4.16 5.01Tb 0.63 1.01 1.03 1.00 0.78 0.79 0.82 0.83 0.65 0.65 0.79Dy 3.82 6.20 6.42 6.01 4.90 4.79 5.00 4.89 3.88 3.97 4.82Ho 0.76 1.17 1.21 1.14 0.94 0.92 0.95 0.95 0.72 0.74 0.91Er 1.82 2.88 3.03 2.82 2.30 2.28 2.28 2.28 1.74 1.84 2.24Tm 0.25 0.39 0.41 0.38 0.31 0.30 0.30 0.30 0.23 0.24 0.30Yb 1.46 2.15 2.30 2.16 1.79 1.72 1.75 1.74 1.31 1.37 1.71Lu 0.22 0.32 0.33 0.31 0.25 0.25 0.25 0.27 0.19 0.20 0.24Ba 148 39 63 37 217 220 233 228 197 201 224Th 1.09 0.73 0.68 0.66 1.33 1.35 1.40 1.38 1.21 1.25 1.38Nb 13.31 12.35 11.36 11.37 19.43 19.59 20.44 19.49 17.55 18.23 19.89Y 19.43 29.75 31.61 29.18 23.87 23.03 24.20 23.49 18.57 19.19 23.41

VOLGEO 2371 14-5-02


One glass analysis (JS-1 in Table 2) of materialfrom the ash deposit (provided to us by JohnSinton) from the cli¡ south of Kalaupapa indi-cates that the material is an alkalic basalt havinghigher FeO, TiO2, K2O, P2O5, and lower Al2O3and SiO2 than the glass from Kalaupapa pillowso¡shore. Although no glass analyses are availablefrom the peninsula itself (despite a speci¢c e¡ortto locate glassy £ow surfaces there), the ash ma-terial has similar chemical characteristics to thecrater wall material. The ash on the cli¡s is chemi-cally similar to the late material erupted from thecrater; it may have erupted at about the time theKalaupapa shield eruption ended.The single whole-rock analysis from the north

submarine £ank of the Kalaupapa shield (Fig. 12)is similar to the non-crater material erupted andcollected on land. The pillow lava glass analyses

of samples collected from 3580 to 390 m depthall contain low S indicating that they were de-gassed on the surface after lava was eruptedfrom a subaerial vent before £owing into the seawhere it chilled to produce pillows and hyaloclas-tite. The glass inclusions in olivine also generallyhave low S (nine of 12 have between 170 and 320ppm S), but three inclusions have S ranging from700 to 840 ppm. The low S contents indicate thatmost of the olivine crystallized and trapped glassduring and after low-pressure (subaerial) degass-ing occurred or after the shield had grown abovesea level. We conclude that submarine eruptedlavas forming much of the volume of the Kalau-papa shield have been completely buried beneathlater subaerially erupted lavas. We also infer thatmost, if not all, the olivine in these lavas formedat very low pressure, perhaps within a pond in the

Table 1A (continued).Whole-rock chemical analyses

Sample: P252-2 P253-2 P253-3 P253-4 P253-8 P253-9 P253-10A P253-10A P253-11A P253-11B P253-12Depth (m): 3560 31755 31680 31615 31450 31350 31280 (Repeat) 31245 31245 31205

ICP-MS analyses (continued)Hf 1.93 3.94 3.93 3.84 2.80 2.76 2.88 2.84 2.42 2.41 2.85Ta 0.78 0.89 0.79 0.81 1.22 1.23 1.30 1.28 1.11 1.13 1.22U 0.37 0.13 0.23 0.20 0.36 0.47 0.47 0.48 0.60 0.53 0.60Pb 0.87 0.91 0.77 0.86 0.99 1.03 1.17 1.22 0.89 1.07 1.17Rb 6.4 2.2 2.4 2.6 9.6 10.2 11.0 10.9 8.3 9.1 10.9Cs 0.12 0.06 0.02 0.01 0.17 0.25 0.20 0.20 0.12 0.14 0.20Sr 285 301 316 297 360 362 395 391 328 333 375Sc 29.4 39.3 40.5 39.1 34.5 33.8 35.3 32.8 27.2 29.5 33.0

Major element analyses by XRF with H2O, FeO, and CO2 determined by wet chemical techniques at the U.S. Geological Survey,Denver, CO. Trace element analyses from the Geoanalytical Laboratory at Washington State University.

Table 1BWhole-rock analyses done at the Geoanalytical Laboratory at Washington State University

Sample: P252-2 P253-2 P253-3 P253-4 P253-8 P253-9 P253-10 P253-11A P253-11B P253-12

SiO2 46.30 49.35 47.83 48.85 47.10 47.05 47.16 45.19 44.60 46.80Al2O3 12.46 13.26 14.31 13.73 15.55 14.93 15.57 11.63 11.45 14.93FeO* 12.86 12.21 12.13 12.23 11.33 11.16 11.45 12.01 12.81 11.54MnO 0.18 0.18 0.17 0.19 0.17 0.16 0.25 0.17 0.17 0.15MgO 13.60 8.52 7.60 8.02 6.75 8.42 6.43 16.32 16.08 8.18CaO 9.96 11.01 11.77 11.23 12.79 12.29 12.85 9.85 9.69 12.31Na2O 2.42 2.24 2.43 2.10 3.01 2.99 3.03 2.32 2.34 2.92K2O 0.33 0.23 0.28 0.30 0.64 0.61 0.64 0.56 0.56 0.66P2O5 0.18 0.24 0.25 0.22 0.22 0.21 0.23 0.22 0.22 0.23TiO2 1.43 2.61 2.62 2.67 2.06 1.99 2.06 1.70 1.69 2.00LOI 30.29 0.46 0.82 0.38 0.28 0.26 0.59 0.20 0.16 0.54Total 99.72 99.85 99.40 99.56 99.62 99.80 99.67 99.97 99.61 99.72

VOLGEO 2371 14-5-02


crater in the shield or during sur¢cial emplace-ment.

7. Discussion

7.1. Reconstructing the shoreline before thelandslide occurred

The lavas on the edge of the break-in-slope aresubaerially erupted and emplaced tholeiitic lava£ows, supporting the interpretation of Mooreand Clague (1992) that the break-in-slope repre-sents the shoreline at the end of voluminous shieldbuilding. The depth to the break-in-slope changesfrom about 31125 m north of the western end ofMolokai to about 31500 m northeast of the east-ern end of Molokai. Another deeper break-in-slope occurs at about 32500 m in the northeast-ern corner of Fig. 4, but it is probably related toKoolau Volcano on Oahu since it is truncated bythe slope and break-in-slope of West Molokai.Within the region of canyons north of East Mo-lokai, the break-in-slope depth is not as sharp aboundary as either east or west of this region. Thecanyons have cut into the break-in-slope andmade it irregular in plan view and in depth, butit is still at about 31300 m depth.If the landslide occurred after the shoreline had

stabilized at the end of voluminous shield-build-

ing, we would expect the break-in-slope within theheadwall to have retrogressed towards the islandand the depth to the break-in-slope should beshallower within the headwall region. This is notthe case, suggesting that the shoreline continuedto be rebuilt by lava £ows after the landslide oc-curred and the headwall formed. It is also possiblethat the landslide headwall was farther o¡shore,beyond the break-in-slope, when the slide oc-curred. Since no additional scarp is seen o¡shore,the headwall would then have had to retrogress tothe shoreline. Such a coincidence seems unlikely,so we conclude that the original shoreline wasbowed outward from the island as a largebench-like structure, and that the landslide failurea¡ected only the submarine portion of this bench.Subsequent to the landslide, East Molokai Volca-no rebuilt the shoreline break-in-slope, secondaryslumping retrogressed the crown scarp (displacingthe block shown in Fig. 10), the main episodeof shield-building volcanism slowed and thenstopped, erosion cut canyons to the shoreline,and the area subsided by about 1500 m. Immedi-ately upon subsidence, £ow of meteoric waterdown the subaerial canyons fed subaqueous den-sity £ows, and the work of canyon cutting con-tinued both above and below sea level. Seismicrecords o¡shore northeast Oahu show that thecanyons are sediment ¢lled in the region betweenthe submarine and subaerial canyons (Andrews

Table 2Microprobe analyses of glass

Sample P252-1 P252-2 P252-3 P252-9 P253-1 P253-9 P253-11A P253-11B P253-12 JS-1Depth (m) 3580 3560 3475 390 31800 31355 31245 31245 31205 V100# in average 16 16 36 20 20 28 16 16 12 23

SiO2 47.6 47.8 47.9 47.5 51.2 45.8 45.9 45.9 45.9 45.5Al2O3 14.5 14.7 14.4 14.4 13.5 15.2 15.1 15.2 15.2 14.8FeO* 11.8 11.9 12.1 12.2 11.4 12.0 12.2 12.0 11.9 13.1MnO 0.16 0.15 0.15 0.15 0.15 0.16 0.16 0.14 0.14 0.14MgO 6.27 6.44 6.34 6.27 6.50 6.38 6.14 6.16 6.56 5.87CaO 12.4 12.2 12.1 11.9 11.2 12.4 12.6 12.5 12.2 12.9Na2O 3.18 3.16 3.27 3.33 1.89 3.19 3.28 3.30 3.24 2.57K2O 0.37 0.36 0.35 0.37 0.29 0.48 0.56 0.53 0.46 1.31TiO2 1.79 1.74 1.83 1.89 2.62 1.93 2.12 2.08 1.85 3.09P2O5 0.24 0.24 0.26 0.29 0.26 0.22 0.26 0.26 0.22 0.51Cl 0.018 0.022 0.023 0.026 0.007 0.030 0.034 0.034 0.028 0.053S 0.021 0.021 0.018 0.016 0.006 0.100 0.097 0.100 0.102 0.026Total 98.4 98.9 98.8 98.4 99.0 98.1 98.6 98.3 97.9 99.8

VOLGEO 2371 14-5-02


Table 3Microprobe analyses of glass inclusions in olivine

Sample: P252-1a P252-1b P252-1c P252-2a P252-2b P252-2c P252-2d P252-3a P252-3bDepth (m): 3580 3580 3580 3560 3560 3560 3560 3475 3475# in average: 4 4 4 2 4 4 4 4 4

SiO2 47.7 47.9 47.7 47.9 47.8 47.8 47.6 47.9 48.4Al2O3 15.2 15.3 14.5 14.9 14.9 15.2 15.9 14.5 16.1FeO* 11.5 11.3 12.1 12.3 11.9 12.1 10.6 12.3 10.7MnO 0.15 0.16 0.17 0.16 0.16 0.15 0.13 0.16 0.12MgO 6.06 4.76 6.28 6.11 6.36 6.31 6.47 6.34 6.17CaO 12.4 12.9 12.4 12.4 12.2 12.2 12.5 12.1 11.8Na2O 3.18 3.33 3.17 3.16 3.19 3.20 3.28 3.28 3.47K2O 0.36 0.38 0.38 0.38 0.37 0.35 0.35 0.36 0.34TiO2 1.74 1.85 1.84 1.72 1.79 1.69 1.67 1.85 1.63P2O5 0.23 0.24 0.24 0.24 0.24 0.24 0.23 0.29 0.27Cl 0.019 0.026 0.020 0.022 0.021 0.021 0.025 0.026 0.023S 0.018 0.022 0.017 0.022 0.028 0.032 0.070 0.017 0.084Total 98.5 98.1 98.8 99.2 99.0 99.3 98.9 99.1 99.2

Sample: P252-3c P252-9a P252-9b P253-1a P253-1b P253-1c P253-9a P253-9b P253-9cDepth (m): 3475 390 390 31800 31800 31800 31350 31350 31350# in average: 4 4 4 4 4 4 4 4 4


Sample: P253-11Aa P253-11Ab P253-11Ac P253-11Ba P253-11Bb P253-11Bc P253-12a P253-12b P253-12cDepth (m): 31245 31245 31245 31245 31245 31245 31205 31205 31205# in average: 4 4 4 2 4 4 4 4 4


VOLGEO 2371 14-5-02


and Bainbridge, 1972). This ¢lling of the upperpart of the submarine canyons probably occurredduring Pleistocene sea level low stands. The sub-sequent rise in sea level at the end of the last lowstand occurred rapidly and the sediment has notbeen removed from the canyons.

7.2. Formation and history of the northern sea cli¡on Molokai

The sea cli¡ that marks the northern edge ofEast Molokai has been an enigmatic feature. Var-ious authors have attributed its formation to waveerosion and retreat (Dana, 1890; Wentworth,1927; Macdonald and Abbott, 1970) or to a nor-mal fault that down-dropped the block north ofthe coast (Stearns and Macdonald, 1947; alsoproposed in a general way by Dana, 1890). Ithas been proposed as the location of the headwallto the landslide as well (Satake and Smith, 2000).The sea cli¡ is located about 15 km shoreward ofthe coast when voluminous shield constructionceased. There is no doubt that the sea cli¡ hasretreated to the south, some 500 m since forma-tion of the Kalaupapa Peninsula at 330 ka (seeSection 7.4) and, assuming the same rates of re-treat, only about 2.25 km since the end of shieldbuilding on east Molokai about 1.5 Ma. Thus, theproto-sea cli¡ would have been located about 12km inland from the shoreline at the end of shield-building. It therefore seems unlikely that the seacli¡ formed solely by retreat of the shoreline bywave cutting and we must search for an alterna-tive explanation for its formation.The subaerial intercanyon constructional 2^6‡

slopes on East Molokai, if extrapolated north-ward, do not intersect the bottom above thebreak-in-slope. This suggests that the slope be-tween the present shoreline and the break-in-slopemarking the shoreline at the end of shield-build-ing was steeper, on average, than the present sub-aerial constructional slopes. One way, althoughcertainly not the only way, to increase the averageslope is to downfault the northern area. A similarsteep cli¡, the Hilina Pali, occurs on the south£ank of Kilauea Volcano, about halfway betweenthe rift zones and the coastline. Like the northernMolokai sea cli¡, the Hilina Pali decreases in

height from the center towards both ends. Alsolike the northern Molokai sea cli¡, the highestpart of the Hilina Pali is more than 400 m high.On Molokai, the sea cli¡, prior to retreat due toerosion, was located about 0.6 km north of therift zone and 18 km south of the shoreline, where-as the Hilina Pali is located about 5 km southeastof the rift zone and 7 km northwest of the shore-line. Studies of the Hilina Pali demonstrate that itis an active normal or listric fault (Swanson et al.,1976; Hill and Zucca, 1987). Based on the abovearguments, we propose that the north Molokaisea cli¡ formed as a listric fault. Such a faultcould have developed prior to the Wailau land-slide, like the Hilina faults, or it could be a con-sequence of the Wailau slide. Using the south£ank of Kilauea as a model, we suspect that itformed before the landslide, and speculate thatthe throw on the fault may have increased dueto retrogressive slumping after the Wailau land-slide occurred. The general uniform terrace deptheast and west of this proposed listric fault systemrequires that sedimentation on the terrace has¢lled lows and smoothed the terrace after faulting.

7.3. Timing of the Wailau landslide

The age of the Wailau landslide can be roughlyconstrained by the stratigraphic relations of theWailau or Nuuana landslides and the emplace-ment of £ows in the submarine North Arch Vol-canic Field (Clague et al., 1990; Dixon et al.,1997; Clague et al., 2002). Unfortunately, it isunclear which slide is sandwiched between the£ows and the ages of the £ows are still poorlydetermined, although age estimates for the £owsabove and below the landslide blocks are 0.5 and1.6 Ma (Dixon et al., 1997).The on-land geology of East Molokai may pro-

vide additional insight into the timing of the Wai-lau landslide. The two dike swarms on the westernpart of the shield (see Fig. 2) contain basalt withdi¡erent compositions. The northern swarm con-sists of tholeiitic basalt characteristic of the shieldstage and the southern swarm consists of transi-tional basalt that are compositionally similar tothe lavas exposed in the northern sea cli¡ (Beeson,1976; Clague and Beeson, 1980). Since the transi-

VOLGEO 2371 14-5-02


tional lavas of the northern sea cli¡ postdate themain part of the shield stage, we can infer that thenorthern tholeiitic dike swarm is older than thesouthern transitional one. The location of Hawai-ian rift zones is controlled by gravitational stress-es (Fiske and Jackson, 1972), which re£ect thedistribution of mass that forms the volcano. Bee-son (personal communication, 1995) suggestedthat the shift from the northern to the southerndike swarm took place in response to the loss ofmass from the northern £ank of the volcano whenthe Wailau landslide occurred. If Beeson’s explan-ation for the dual dike swarms is correct, then thetiming of the Wailau landslide is constrained atslightly before 1.5 Ma. This age estimate is alsoconsistent with the analysis of the o¡shore break-in-slope, which suggests that the shoreline contin-ued to be rebuilt by lava £ows after the landslideoccurred. Clague et al. (2002) arrived at the same

age based on the geochemistry of glass fragmentsin volcaniclastic rocks contained in some of themore distal Wailau landslide blocks. In particular,they report several glass fragments that are sub-aerially erupted transitional basalt similar to thatexposed in the Kalaupapa cli¡ section describedby Beeson (1976), implying that the slide occurrednear the start of the transition from tholeiiticshield to alkalic postshield volcanism on East Mo-lokai.

7.4. Post-slide volcanism

The Kalaupapa rejuvenated-stage shield formedafter the northern sea cli¡ and the submarine can-yons had largely formed. Likewise, the submarinerejuvenated-stage vents formed after the break-in-slope had submerged. The vents for the Kalaupa-pa shield and the pyroclastic deposit at the base

Fig. 13. Plot of Zr/Y vs. Zr/Nb showing rejuvenated-stage lavas and tholeiitic lava from Molokai, with Kalaupapa Basalt forcomparison and ¢elds for Honolulu Volcanics (Clague and Frey, 1982), Koloa Volcanics (Clague and Dalrymple, 1988), andNorth Arch lavas (Frey et al., 2000). The Kalaupapa and submarine rejuvenated cone from north of Molokai share low Zr/Nbwith all other rejuvenated-stage lavas, but are at the low end of the range of Zr/Y, suggesting less residual garnet is left in theirsource compared to most other rejuvenated-stage lavas. The tholeiitic lavas from the break-in-slope have Zr/Nb between that forKilauea and Mauna Loa (Rhodes et al., 1989), but slightly lower Zr/Y, again suggesting less residual garnet in their source re-gion.

VOLGEO 2371 14-5-02


of the cli¡ are both located on or near the normalfault that formed the sea cli¡. The o¡shore ventsare located between the Hilina-like fault thatmarks the northern sea cli¡ and the break-in-slope that roughly locates the headwall of theslide.The lack of any but subaerial pahoehoe lavas

on the Kalaupapa Peninsula and the presence ofpillow lava up to as shoal as 345 m indicates thatpaleo-sea level when the peninsula formed wasbetween 345 m and present-day sea level. Anoth-er line of evidence pointing to minimal post-Ka-laupapa subsidence is the notable slope change onthe shield at about 340 m depth (Fig. 14A). Thisabrupt steepening of the slope is similar to thatwhich occurs on the active Hawaiian volcanoesand apparently represents the position of sea levelat the time of shield formation (Moore, 1987) ingeneral agreement with the subaqueous observa-tions. Sea level only reached these levels from 320to 340 ka, from 390 to 410 ka, and from 490 to505 ka within the age range de¢ned by the KÂrages (Fig. 15). The highest sea level, and one thatwould also accommodate modest subsequent is-land subsidence, occurred between 320 and 340ka and is at the margin of error of two of thethree KÂr ages. The 390^410-ka and the 490^505-ka possible ages coincide with high sea leveland could allow for even more modest island sub-sidence, although these ages fall between the ex-isting KÂr ages. If the Kalaupapa shield formedat 490^505 ka, then Molokai has subsided at lessthan 0.03 mm/yr in the past 500 ka. This analysissuggests that the Kalaupapa Peninsula and theisland of Molokai have undergone no more than30 m of subsidence in the 330 ka since the pen-insula was most likely built, consistent with a sub-sidence rate less than 0.1 mm/yr.

Fig. 14. Pro¢les of north slope of Molokai. (A) Four N^Spro¢les covering an area 18 km wide including island and ex-tending north to about 2500 m depth. Arrows show positionof two submersible dives. The Kalaupapa volcano is crossedby pro¢le along 156‡58.7PN Lat. (B) Six N^S pro¢les cover-ing an area 43 km wide extending north to 300 m depth.The Kalaupapa volcano is not crossed. (C) Depth pro¢lesfrom east to west showing the depth to the 3140 m terrace,and the top and bottom of the major break-in-slope at31200 to 31500 m.

VOLGEO 2371 14-5-02


7.5. Subsidence history of Molokai

The reconstruction of sea level during thelikely time that the Kalaupapa shield formed(Section 7.4) suggests that Molokai has subsidedless than 30 m, and perhaps not at all, in the last0.3^0.4 Ma. The 3140 m terrace along the re-mainder of the north slope of Molokai wasprobably formed of reef and deltaic deposits, pri-marily after subsidence abated, during the numer-ous Pleistocene glacial maxima low stands ofthe sea that occurred in the last 700 ka (see Fig.15).The submarine break-in-slope is also a shore-

line marker at the end of voluminous shield-build-ing. On West Molokai, this shoreline occurs at31125 to 31300 m depth, tipping down to theeast. This amount of subsidence has taken placesince about 1.9 Ma and the tipping indicates morerapid subsidence to the east during the periodafter West Molokai had largely been built (after

1.9 Ma) and until East Molokai Volcano waslargely built (by 1.5 Ma).On East Molokai, the Wailau landslide head-

wall disrupts much of this original break-in-slope,but east of the landslide headwall, the break-in-slope is about 31500 m deep, also tipping downto the east. This amount of subsidence has takenplace in the past 1.5 Ma and the eastward tippingis caused by the subsequent construction of WestMaui Volcano, and particularly the larger EastMaui (Haleakala) Volcano, which £exed the litho-sphere down to the east.Since no signi¢cant subsidence has occurred in

the past 0.3^0.4 Ma, we can calculate averagesubsidence rates of 0.8 mm/y (V1200 m between1.9 and 0.35 Ma) and 1.3 mm/yr (V1500 m be-tween 1.5 and 0.35 Ma) for West and East Mo-lokai, respectively. As Fig. 6 shows, however, alarge part of the subsidence takes place whilethe volcanoes are still active and lava £ows occa-sionally cross the shoreline and construct lava del-

Fig. 15. Pleistocene sea level curve (based on the Ontong^Java Plateau record in Berger et al., 1993) showing likely times thatthe Kalaupapa shield formed. Subsidences of 0, 0.1, and 0.05 mm/yr are shown as lines projecting back in time from the present340 m depth of the break-in-slope on the £anks of the Kalaupapa Peninsula. KÂr ages, with analytical errors, are shown fromClague et al. (1982).

VOLGEO 2371 14-5-02


tas. This observation suggests that at least 500 mof the total 1200 m of subsidence of West Molo-kai took place in only the ¢rst V0.2 Ma follow-ing the end of voluminous shield-building. Thisrate (2.5 mm/yr) is identical to that of the muchlarger Hawaii (Clague and Moore, 1991; Mooreand Clague, 1992) and suggests that subsidenceoccurs at similar rates for small and large volca-noes, but that, since total subsidence is larger forlarger volcanoes, the subsidence simply continuesthere for longer time periods. At their zenith,West and East Molokai Volcanoes stood roughly1.6 and 3 km above sea level, respectively.

7.6. Relation of canyons and landslides

Moore et al. (1989) noted that submarine can-yons commonly are associated with large land-slides in Hawaii. A causal relation has been in-ferred, but it has been unclear whether thelandslides cause the canyons to form by over-steepening the slope of the volcano, or if the pres-ence of the canyons somehow leads to slope fail-ure. Molokai is one of the many places in Hawaiiwhere canyons occur on the slope above a largecatastrophic landslide. At roughly the time theWailau landslide occurred, East Molokai volcanostood 3.0 km above sea level causing subaerialstreams on the windward side to cut giant can-yons down to the shoreline (at the present-dayV1500 m depth level). The subaerial cutting ofthe upper submarine canyons indicates high sur-face runo¡ and therefore high rainfall in the peri-od during and after the shield-stage ended, andwhile the volcano was still subsiding fairly rapidly(probably from 1.5 to roughly 1 Ma). The sub-marine erosion of the canyons, following theirsubmergence, and headward erosion of the can-yons at the break-in-slope suggest continuing sub-marine £ow of sediment-laden density currents-again attesting to high surface runo¡ from Molo-kai Island up to the present time.Deterich (1988) and Ellsworth and Voight

(1996) evaluated the role of dike injection andIverson (1995) analyzed the interplay of magmainjection and groundwater forces in causing the

large Hawaiian landslides. Iverson (1995) con-cluded that a thick impermeable clay layer wasrequired to generate head gradients that couldtrigger such large landslides but that it seemedunlikely that such a low-friction unit of clay couldbe present or that ground-water forces could re-duce friction enough to trigger the slides. Clagueand Denlinger (1994) suggested that £ow of acumulate dunite mass within the volcano couldadd to the magma injection forces and help desta-bilize the £anks of Hawaiian volcanoes. Magmasystem forces associated with dike injection andcumulate £ow are su⁄cient to cause spreading ofthe volcano £anks (Borgia and Treves, 1992) andconsequent formation of a bench like that on Ki-lauea’s south £ank today (Smith et al., 1999), butadditional forces may be required to trigger thelarge catastrophic slides. Clague and Dixon (2000)synthesized data suggesting that extrinsic varia-bles, such as rainfall, have profound e¡ects onthe evolution of volcanic systems in Hawaii.Groundwater pressurization caused by magma in-trusion could be such a force. If this is the case,then availability of groundwater near the magmachamber could be the key to landslide triggeringand high rainfall could be an important variablethat determines whether the spreading £ank of avolcano eventually fails catastrophically. Avail-ability of surface water or groundwater near themagma system is also required for phreatomag-matic eruptions (Mastin, 1997). Phreatomagmaticeruptions could generate large lateral forces thatcontribute to £ank failures, although such forcesshould be concentrated in the shallowest parts ofthe volcano where volume expansion of water tosteam is greatest. How, or if, such forces contrib-ute to deep-seated failures like the Hawaiian giantlandslides remains unclear. However, high rainfallprovides surface runo¡ to carve the canyons,groundwater forces to reduce £ank friction duringdike intrusions, and groundwater that can mixwith magma stored in the crustal magma reservoirto trigger phreatomagmatic eruptions. Althoughthe exact mechanism remains elusive, ground-water appears to be one component needed totrigger the catastrophic landslides.

VOLGEO 2371 14-5-02


8. Conclusions

(1) The main break-in-slope on the north sub-marine £ank of Molokai at 31500 to 31250 m isa shoreline feature that has been only modestlymodi¢ed by the Wailau landslide.(2) The weight of the evidence suggests that the

submarine canyons above the break-in-slope, in-cluding one meandering stream, were initially sub-aerially carved. Where such canyons cross thebreak-in-slope, plunge pools may form by erosioncaused by sediment-laden density currents plung-ing over the cli¡.(3) Continued infrequent volcanic activity near

the end of the shield stage on West Molokai Vol-cano at about 1.8 Ma formed a series of smallcoastal sea cli¡s as the island subsided.(4) Lava £ows exposed at the break-in-slope

and down to 1800 m are subaerially erupted andemplaced tholeiitic shield lava.(5) Submarine rejuvenated-stage volcanic cones

formed after the landslide took place and follow-ing at least 400^500 m of subsidence after themain break-in-slope had formed.(6) The sea cli¡ on east Molokai is not the

headwall of the landslide, nor did it form entirelyby erosion. It may mark the location of a listricfault similar to the Hilina faults on present-dayKilauea Volcano.(7) The Wailau landslide occurred about 1.5 Ma

and the Kalaupapa Peninsula most likely formedat 3307 5 ka, but could alternatively have formedat 390^410 ka or 490^505 ka.(8) Molokai is presently stable relative to sea

level and has subsided no more than 30 m inthe last 330 ka.(9) At their peak, West and East Molokai stood

1.6 and 3 km above sea level.(10) High rainfall causes high surface runo¡

and formation of subaerial canyons, increasesgroundwater forces during dike intrusions thatreduce £ank friction, and provides groundwaterthat can mix with shallow stored magma to trig-ger phreatomagmatic eruptions. Active volcanismand high rainfall appear to be two componentsneeded to trigger the deep-seated giant Hawaiianlandslides.

Acknowledgements

We thank Jennifer Reynolds, Norm Maher,Gerry Hatcher and Bill Danforth for theirassistance at sea collecting and processing theSimrad EM300 bathymetric and backscatter dataand Dave Caress for his subsequent assistance infurther processing the data on shore. The Pisces Vdives were supported by the NOAA HawaiiUndersea Research Program and their successwas due to the dedication of the ship andsubmersible crew, particularly chief pilot TerryKirby. We also thank John Smith for ablysubstituting at the last moment as a diver on diveP253 and John Sinton for kindly sending us asample of the pyroclastic deposit from the base ofthe cliff on East Molokai. Mel Beeson mademicroprobe analyses of glass and provided valu-able insight into the setting of Molokai’s riftzones. Derek Elsworth and an anonymous re-viewer provided helpful comments.

References

Andrews, J.E., Bainbridge, C., 1972. Submarine canyons o¡eastern Oahu. Pac. Sci. 26, 108^113.

Beeson, M.H., 1976. Petrology, mineralogy, and geochemistryof the East Molokai Volcanic Series. U.S. Geol. Surv. Prof.Paper 961, 53 pp.

Berger, W.H., Kroenke, L.W., Mayer, L.A., et al., 1983. Proc.ODP, Sci. Results, 130, 867 pp.

Borgia, A., Treves, B., 1992. Volcanic plates overriding theoceanic crust: structure and dynamics of Hawaiian volca-noes. Geol. Soc. London Spec. Publ. 60, 277^299.

Clague, D.A., 1987a. Petrology of West Molokai Volcano.Geol. Soc. Am. Abstr. 19, 366.

Clague, D.A., 1987b. Hawaiian alkaline volcanism. In: Fitton,J.G., Upton, B.G.J. (Eds.), Alkaline Igneous Rocks. Geo-logical Society Special Publication No. 30, pp. 227^252.

Clague, D.A., Beeson, M.H., 1980. Trace element geochemis-try of the East Molokai Volcanic Series, Hawaii. Am. J. Sci.280-A, 820^844.

Clague, D.A., Dalrymple, G.B., 1988. Age and petrology ofalkalic postshield and rejuvenated-stage lava from Kauai,Hawaii. Contrib. Mineral. Petrol. 99, 202^218.

Clague, D.A., Denlinger, R.P., 1994. The role of olivine cu-mulates in destabilizing the £anks of Hawaiian volcanoes.Bull. Volcanol. 56, 425^434.

Clague, D.A., Dixon, J.E., 2000. Extrinsic controls on theevolution of Hawaiian ocean island volcanoes, Geochem.Geophys. Geosyst. 1, 12 pp.

VOLGEO 2371 14-5-02


Clague, D.A., Frey, F.A., 1982. Petrology and trace-elementgeochemistry of the Honolulu Volcanics, Oahu: Implica-tions for the oceanic mantle below Hawaii. J. Petrol. 23,447^504.

Clague, D.A., Moore, J.G., 1991. Geology and petrologyof Mahukona volcano, Hawaii. Bull. Volcanol. 53, 159^172.

Clague, D.A., Dao-gong, C., Murnane, R., Beeson, M.H.,Lanphere, M.A., Dalrymple, G.B., Friesen, W., Holcomb,R.T., 1982. Age and petrology of the Kalaupapa Basalt,Molokai, Hawaii. Pac. Sci. 36, 411^420.

Clague, D.A., Frey, F.A., Beeson, M.H., 1983. Rare-earth-el-ement and Sr-isotopic evidence for the origin of the EastMolokai Volcanics, Hawaii. EOS, Trans. Am. Geophys.Union 64, 902.

Clague, D.A., Holcomb, R.T., Sinton, J.M., Detrick, R.S.,Torresan, M.E., 1990. Pliocene and Pleistocene alkalic £oodbasalts on the sea£oor north of the Hawaiian Islands. EarthPlanet. Sci. Lett. 98, 175^191.

Clague, D.A., Moore, J.G., and Davis, A.S., 2002. Formationof volcanic breccia and hyaloclastite in blocks from the Wai-lau and Nuuanu giant landslides, Hawaii. In: Hawaiian Vol-canoes: Deep Underwater Perspectives. Am. Geophys.Union Monogr. 128, 279-296.

Coombs, C.R., Hawke, B.R., Wilson, L., 1990. Terrestrialanalogs to Lunar sinuous rilles: Kauhako Crater and chan-nel. Kalaupapa, Molokai, and other Hawaiian lava conduitsystems. Proc. 20th Lunar and Planet. Sci. Conf., Lunar andPlanet. Inst., Houston, TX, pp. 195^206.

Coulbourn, W.T., Campbell, J.F., Moberly, R., 1974. Hawai-ian submarine terraces, canyons, and Quaternary historyevaluated by seismic-re£ection pro¢ling. Mar. Geol. 17,215^234.

Dana, J.D., 1890. Characteristics of Volcanoes. UniversityPress, Cambridge, MA, 399 pp.

Dartnell, P., Gardner, J.V., 1999. Sea-£oor images and datafrom multibeam surveys in San Francisco Bay, SouthernCalifornia, Hawaii, the Gulf of Mexico, and Lake Tahoe,California-Nevada. U.S. Geol. Surv. Digital Data SeriesDDS-55.

Deterich, J.H., 1988. Growth and persistence of Hawaiian vol-canic rift zones. J. Geophys. Res. 93, 4258^4270.

Dixon, J.E., Clague, D.A., Poreda, Wallace, P., 1997. Volatilesin alkalic basalts from the North Arch Volcanic Field, Ha-waii : Extensive degassing of deep submarine-erupted alkalicseries lavas. J. Petrol. 38, 911^939.

Elsworth, D., Voight, B., 1996. Evaluation of volcano £ankinstability triggered by dyke intrusion. Geol. Soc. Spec.Publ. 110, 45^53.

Fiske, R.S., Jackson, E.D., 1972. Orientation and growth ofHawaiian volcanic rifts ^ the e¡ect of regional structure andgravitational stresses. R. Soc. London, Proc. Ser. A, 329,299^326.

Frey, F.A., Clague, D.A., Mahoney, J.J., Sinton, J.M., 2000.Volcanism at the edge of the Hawaiia plume: petrogenesis ofsubmarine alkalic lavas from the North Arch Volcanic Field.J. Petrol. 41, 667^691.

Hampton, M.A., Lee, H.J., Locat, J., 1996. Submarine land-slides. Rev. Geophys. 34, 33^59.

Hill, D.P., Zucca, J.J., 1987. Geophysical constraints on thestructure of Kilauea and Mauna Loa volcanoes and someimplications for seismomagmatic processes. U.S. Geol. Surv.Prof. Pap. 1350, 903^917.

Holcomb, R.T., 1985. The caldera of East Molokai Volcano,Hawaiian Islands. Natl. Geogr. Soc. Res. Reports 21, 81^87.

Iverson, R.M., 1995. Can magma-injection and groundwaterforces cause massive landslides on Hawaiian volcanoes?J. Volcanol. Geotherm. Res. 66, 295^308.

Langenheim, V.A.M., Clague, D.A., 1987. The Hawaiian-Em-peror volcanic chain, Part II. Stratigraphic framework ofvolcanic rocks of the Hawaiian Islands. U.S. Geol. Surv.Prof. Pap. 1350, 55^84.

Lee, H.J., 1989. Undersea landslides: extent and signi¢cance inthe Paci¢c Ocean. In: Brabb, E.E. (Ed.), Landslides: Extentand Economic Signi¢cance. Proceedings of the 28th Int.Geol. Cong. Symp. on Landslides. Balkema, Rotterdam,pp. 361^374.

Macdonald, G.A., Abbott, A.T. 1970. Volcanoes in the sea,the geology of Hawaii. University of Hawaii Press, Honolu-lu, HA, 441 pp.

Mark, R.K., Moore, J.G., 1987. Slopes of the Hawaiian Ridge.U.S. Geol. Surv. Prof. Pap. 1350, 101^107.

Mastin, L.G., 1997. Evidence for water in£ux from a calderalake during the explosive hydromagmatic eruption of 1790,Kilauea Volcano, Hawaii. J. Geophys. Res. 102, 20093^20109.

Mathewson, C.C., 1970. Submarine canyons and the shelfalong the north coast of Molokai Island, Hawaiian Ridge.Pac. Sci. 24, 235^244.

MBARI Mapping Team, 2000, MBARI Hawaii Multibeamsurvey. Monterey Bay Aquarium Research Institute DigitalData Series No. 2 (2 CD set), Moss Landing, CA.

McDougall, I., 1964. Potassium-argon ages from lavas of theHawaiian Islands. Geol. Soc. Am. Bull. 75, 107^128.

Moore, J.G., 1964. Giant submarine landslides on the Hawai-ian Ridge. U.S. Geol. Surv. Prof. Pap. 501-D, D95^D98.

Moore, J.G., 1987. Subsidence of the Hawaiian Ridge. U.S.Geol. Surv. Prof. Pap. 1350, 85^100.

Moore, J.G., Campbell, J.F., 1987. Age of tilted reef, Hawaii.J. Geophys. Res. 92, 2641^2646.

Moore, J.G., Clague, D.A., 1987. Coastal lava £ows fromMauna Loa and Hualalai Volcanoes, Kona, Hawaii. Bull.Volcanol. 49, 752^764.

Moore, J.G., Clague, D.A., 1992. Volcano growth and evolu-tion of the island of Hawaii. Bull. Geol. Soc. Am. 104,1471^1484.

Moore, J.G., and Clague, D.A., 2002. Mapping the Nuuanuand Wailau landslides. In: Hawaiian Volcanoes: DeepUnderwater Perspectives. Am. Geophys. Union Monogr.128, 223-244.

Moore, J.G., Clague, D.A., Holcomb, R.T., Lipman, P.W.,Normark, W.R., Torresan, M.E., 1989. Prodigious submar-ine landslides on the Hawaiian Ridge. J. Geophys. Res. 94,17,465^17,484.

VOLGEO 2371 14-5-02


Moore, J.G., Normark, W.R., Holcomb, R.T., 1994a. GiantHawaiian landslides. Annu. Rev. Earth Sci. 22, 119^144.

Moore, J.G., Normark, W.R., Holcomb, R.T., 1994b. GiantHawaiian underwater landslides. Science 264, 46^47.

Moore, J.G., Thomas, D.M., 1988. Subsidence of Puna, Ha-waii inferred from sulfur content of drilled lava £ows.J. Volcanol. 35, 165^171.

Naka, J., Takahashi, E., et al., 2000. Tectono-magmatic pro-cesses investigated at deep-water £anks of Hawaiian volca-noes. EOS, Trans. Am. Geophys. Union 81, 221 and 226^227.

Naughton, J.J., Macdonald, G.A., Greenberg, V.A., 1980.Some additional potassium-argon ages of Hawaiian rocks:the Maui volcanic complex of Molokai, Lanai, and Maui.J. Volc. and Geotherm. Res. 7, 339^355.

Rhodes, J.M., Wenz, K.P., Neal, C.A., Sparks, J.W., Lock-wood, J.P., 1989. Geochemical evidence for invasion ofKilauea’s plumbing system by Mauna Loa magma. Nature337, 257^260.

Satake, K., Smith, J.R., 2000. Tsunami modeling from Hawai-ian submarine landslides. EOS, Trans. Am. Geophys. Union81, WP251.

Shepard, F.P., Dill, R.F., 1966. Submarine Canyons and OtherSea Valleys. Rand McNally, Chicago, IL, 381 pp.

Smith, J.R., Malaho¡, A., Shor, A.N., 1999. Submarine geol-ogy of the Hilina slump and morpho-structural evolution ofKilauea Volcano, Hawaii. J. Volcanol. Geotherm. Res. 94,59^88.

Smith, J.R., Satake, K., Suyehiro, K., 2002. Deepwater Multi-beam Sonar Surveys Along the Southeastern Hawaii Ridge:

Guide to the CD-ROM. In: Hawaiian Volcanoes: DeepUnderwater Perspectives. Am. Geophys. Union Monogr.128, 3-9.

Somers, M.L., Carson, R.M., Revie, J.A., Ege, R.H., Barrow,V.J., Andrews, A.G., 1978. GLORIA II ^ An improved longrange sidescan sonar. In: Proceedings, IEE/IERE Subcon-ference on Ocean Instruments and Communication, Ocean-ology International. B.P.S., London, pp. 16^24.

Stearns, H.T., Macdonald, G.A., 1947. Geology and ground-water resources of the island of Molokai, Hawaii. HawaiiDiv. Hydrogr. Bull. 11, 113 pp.

Swanson, D.A., Du⁄eld, W.A., Fiske, R.S., 1976. Displace-ment of the south £ank of Kilauea volcano: the result offorceful intrusion of magma into the rift zones. U.S. Geol.Surv. Prof. Pap. 963, 30 pp.

Varnes, D.J., 1978. Slope movement types and processes. In:Schuster, R.L., Krizek, R.J. (Eds.), Landslides, Analysis andControl. National Academy of Sciences Spec. Report 176,pp. 11^33.

Walker, G.P.L., 1990. Kalaupapa and Mokuhooniki, Molokai,Hawaii. In: Wood, C.A., Kienle, J. (Eds.), Volcanoes ofNorth America. Cambridge University Press, Cambridge,pp. 330^331.

Wentworth, C.K., 1927. Estimates of marine and £uvial ero-sion in Hawaii. J. Geol. 1, 117^133.

Wilde, P., Chase, T.E., Normark, W.R., Thomas, J.A., Young,J.D., 1980. Oceanographic data o¡ the southern HawaiianIslands. LBL Publ. 359, Lawrence Berkeley Lab., Universityof California, Berkeley, CA.

VOLGEO 2371 14-5-02


Date post:	10-Dec-2023
Category:	Documents
Upload:	independent
View:	0 times
Download:	0 times

The proximal part of the giant submarine Wailau landslide, Molokai, Hawaii

Documents