Geology topics
Peter W. Lipman
Publications and source records attributed to Peter W. Lipman.
Eruptive and noneruptive calderas, northeastern San Juan Mountains, Colorado: Where did the ignimbrites come from?
The northeastern San Juan Mountains, the least studied portion of this well-known segment of the Southern Rocky Mountains Volcanic Field are the site of several newly identified and reinterpreted ignimbrite calderas. These calderas document some unique eruptive features not described before from large volcanic systems elsewhere, as based on recent mapping, petrologic data, and a large array of newly determined high-precision, laser-fusion 40 Ar/ 39 Ar ages (140 samples). Tightly grouped sanidine ages document exceptionally brief durations of 50–100 k.y. or less for individual Oligocene caldera cycles; biotite ages are more variable and commonly as much as several hundred k.y. older than sanidine from the same volcanic unit. A previously unknown ignimbrite caldera at North Pass, along the Continental Divide in the Cochetopa Hills, was the source of the newly distinguished 32.25-Ma Saguache Creek Tuff (~400–500 km 3 ). This regionally distinctive crystal-poor alkalic rhyolite helps fill an apparent gap in the southwestward migration from older explosive activity, from calderas along the N-S Sawatch locus in central Colorado (youngest, Bonanza Tuff at 33.2 Ma), to the culmination of Tertiary volcanism in the San Juan region, where large-volume ignimbrite eruptions started at ca. 29.5 Ma and peaked with the enormous Fish Canyon Tuff (5000 km 3 ) at 28.0 Ma. The entire North Pass cycle, including caldera-forming Saguache Creek Tuff, thick caldera-filling lavas, and a smaller volume late tuff sheet, is tightly bracketed at 32.25–32.17 Ma. No large ignimbrites were erupted in the interval 32–29 Ma, but a previously unmapped cluster of dacite-rhyolite lava flows and small tuffs, areally associated with a newly recognized intermediate-composition intrusion 5 × 10 km across (largest subvolcanic intrusion in San Juan region) centered 15 km north of the North Pass caldera, marks a near- caldera-size silicic system active at 29.8 Ma. In contrast to the completely filled North Pass caldera that has little surviving topographic expression, no voluminous tuffs vented directly from the adjacent Cochetopa Park caldera, which is morphologically beautifully preserved. Instead, Cochetopa Park subsided passively as the >500 km 3 Nelson Mountain Tuff vented at 26.9 Ma from an “underfit” caldera (youngest of the San Luis complex) 30 km to the SW. Three separate regional ignimbrites were erupted sequentially from San Luis calderas within an interval of less than 50–100 k.y., a more rapid recurrence rate for large explosive eruptions than previously documented elsewhere. In eruptive processes, volcanic compositions, areal extent, duration of activity, and magmatic production rates and volumes, the Southern Rocky Mountains Volcanic Field represents present-day erosional remnants of a composite volcanic field, comparable to younger ignimbrite terranes of the Central Andes.
Incremental assembly and prolonged consolidation of Cordilleran magma chambers--Evidence from the Southern Rocky Mountain volcanic field
Recent inference that Mesozoic Cordilleran plutons grew incrementally during >10 6 yr intervals, without the presence of voluminous eruptible magma at any stage, minimizes close associations with large ignimbrite calderas. Alternatively, Tertiary ignimbrites in the Rocky Mountains and elsewhere, with volumes of 1–5 × 10 3 km 3 , record multistage histories of magma accumulation, fractionation, and solidification in upper parts of large subvolcanic plutons that were sufficiently liquid to erupt. Individual calderas, up to 75 km across with 2–5 km subsidence, are direct evidence for shallow magma bodies comparable to the largest granitic plutons. As exemplified by the composite Southern Rocky Mountain volcanic field (here summarized comprehensively for the first time), which is comparable in areal extent, magma composition, eruptive volume, and duration to continental-margin volcanism of the central Andes, nested calderas that erupted compositionally diverse tuffs document deep composite subsidence and rapid evolution in subvolcanic magma bodies. Spacing of Tertiary calderas at distances of tens to hundreds of kilometers is comparable to Mesozoic Cordilleran pluton spacing. Downwind ash in eastern Cordilleran sediments records large-scale explosive volcanism concurrent with Mesozoic batholith growth. Mineral fabrics and gradients indicate unified flow-age of many pluton interiors before complete solidification, and some plutons contain ring dikes or other textural evidence for roof subsidence. Geophysical data show that low-density upper-crustal rocks, inferred to be plutons, are 10 km or more thick beneath many calderas. Most ignimbrites are more evolved than associated plutons; evidence that the subcaldera chambers retained voluminous residua from fractionation. Initial incremental pluton growth in the upper crust was likely recorded by modest eruptions from central volcanoes; preparation for caldera-scale ignimbrite eruption involved recurrent magma input and homogenization high in the chamber. Some eroded calderas expose shallow granites of similar age and composition to tuffs, recording sustained postcaldera magmatism. Plutons thus provide an integrated record of prolonged magmatic evolution, while volcanism offers snapshots of conditions at early stages. Growth of subvolcanic batholiths involved sustained multistage open-system processes. These commonly involved ignimbrite eruptions at times of peak power input, but assembly and consolidation processes continued at diminishing rates long after peak volcanism. Some evidence cited for early incremental pluton assembly more likely records late events during or after volcanism. Contrasts between relatively primitive arc systems dominated by andesitic compositions and small upper-crustal plutons versus more silicic volcanic fields and associated batholiths probably reflect intertwined contrasts in crustal thickness and magmatic power input. Lower power input would lead to a Cascade- or Aleutian-type arc system, where intermediate-composition magma erupts directly from middle- and lower-crustal storage without development of large shallow plutons. Andean and southern Rocky Mountain–type systems begin similarly with intermediate-composition volcanism, but increasing magma production, perhaps triggered by abrupt changes in plate boundaries, leads to development of larger upper-crustal reservoirs, more silicic compositions, large ignimbrites, and batholiths. Lack of geophysical evidence for voluminous eruptible magma beneath young calderas suggests that near-solidus plutons can be rejuvenated rapidly by high-temperature mafic recharge, potentially causing large explosive eruptions with only brief precursors.
Time scales and volumes of large ignimbrite-caldera eruptions in continental arc: Relation to assembly of subvolcanic batholiths
Volcanoes and upper-crustal plutons in diverse geologic settings tend to share common features of mineral and chemical compositions, emplacement age, and magmatic volume. Voluminous silicic ignimbrites associated with caldera sources, widespread components of Cordilleran arcs, have commonly been interpreted as broadly concurrent with assembly of upper-crustal batholiths. Tertiary ignimbrites in the westerns USA and elsewhere, with volumes to 1-5x10 3 km 3 , record multi-stage histories of magma accumulation, fractionation, and solidification in upper parts of large subvolcanic plutons that were sufficiently liquid to erupt. Indiviudal calderas, to 75 km across with 2-5 km subsidence, are direct evidence for shallow magma bodies comparable to the largest granitic plutons. Nested polycyclic calderas that erupted compositionally diverse tuffs, some with reposed intervals of 100 ka or less, document deep composite subsidence and rapid evolution in subvolcanic magmas. Most ignimbrite compositions are more evolved than associated plutons, requiring that subcaldera chambers retained voluminous residua from fractionation. Geophysical data that shows that low-density upper-crustal rocks, inferred to be plutons, are 10km or more thick beneath many calderas. Alternatively, some recent field and geochronologic studies have been interpreted as indicating that individual Mesozoic Cordilleran plutons grew and solidified incrementally in small batches during > 10 6- year intervals, without presence of voluminous eruptible magma ("large tank") at any stage during pluton growth and batholith assembly. Such growth in plutons in small increments would minimize close associations with large ignimbrite calderas and suggest that batholith growth is largely unrelated to surface volcanism. Linked to these interpretations are inferences that ignimbrite eruptions record ephemeral magma chambers that (1) grow rapidly due to exceptionally high magmatic power input to the upper crust, (2) evacuate nearly completely during ignimbrite eruption, and (3) leave little geologic record in the form of crystallized crustal plutons. How to reconcile these alternatives? Many large continental arcs record a broadly unified time-space-composition evolution of upper-crustal magmatic systems. Such volcanic fields especially those containing ignimbrite-caldera episodes, commonly contain compositionally diverse eruptive products erupted over multimillion-year intervals. A common pattern is initial eruptions of intermediate-composition lavas from central volcanoes, followed by eruption of one or more large-volume ignimbrites of more silicic composition; concurrent caldera subsidence is located centrally within the are of prior lava vents. Such progressions of surface volcanism can be interpreted as providing instantaneous sequential snapshots of changing magma-chamber process through time. In contrast, subvolcanic plutons exposed in eroded volcanic terranes represent time-integrated and partly homogenized end products, as successive magmatic pulses accumulated, fractionated, and consolidated in the upper crust. Such perspectives combine evidence for prolonged growth and incremental pluton assembly with presence of large-volume eruptible chambers during peak magmatic input.
Isotope geochemistry of early Kilauea magmas from the submarine Hilina bench: The nature of the Hilina mantle component
Submarine lavas recovered from the Hilina bench region, offshore Kilauea, Hawaii Island provide information on ancient Kilauea volcano and the geochemical components of the Hawaiian hotspot. Alkalic lavas, including nephelinite, basanite, hawaiite, and alkali basalt, dominate the earliest stage of Kilauea magmatism. Transitional basalt pillow lavas are an intermediate phase, preceding development of the voluminous tholeiitic subaerial shield and submarine Puna Ridge. Most alkalic through transitional lavas are quite uniform in Sr–Nd–Pb isotopes, supporting the interpretation that variable extent partial melting of a relatively homogeneous source was responsible for much of the geochemical diversity of early Kilauea magmas ( Sisson et al., 2002 ). These samples are among the highest 206 Pb/ 204 Pb known from Hawaii and may represent melts from a distinct geochemical and isotopic end-member involved in the generation of most Hawaiian tholeiites. This end-member is similar to the postulated literature Kea component, but we propose that it should be renamed Hilina, to avoid confusion with the geographically defined Kea-trend volcanoes. Isotopic compositions of some shield-stage Kilauea tholeiites overlap the Hilina end-member but most deviate far into the interior of the isotopic field defined by magmas from other Hawaiian volcanoes, reflecting the introduction of melt contributions from both “Koolau” (high 87 Sr/ 86 Sr, low 206 Pb/ 204 Pb) and depleted (low 87 Sr/ 86 Sr, intermediate 206 Pb/ 204 Pb) source materials. This shift in isotopic character from nearly uniform, end-member, and alkalic, to diverse and tholeiitic corresponds with the major increase in Kilauea's magmatic productivity. Two popular geodynamic models can account for these relations: (1) The upwelling mantle source could be concentrically zoned in both chemical/isotopic composition, and in speed/extent of upwelling, with Hilina (and Loihi) components situated in the weakly ascending margins and the Koolau component in the interior. The depleted component could be refractory and spread throughout or scavenged from the overlying lithosphere. (2) The Hilina (and Loihi) components could be a more fertile material (lower melting temperature) spread irregularly throughout the Hawaiian source in a matrix of more refractory depleted and Koolau compositions. Modest upwelling along the leading hotspot margin melts the fertile domains predominantly, while the refractory matrix also partially melts in the more vigorously upwelling hotspot interior, diluting the Hilina and Loihi components and yielding voluminous isotopically diverse tholeiitic magmas.
Piggyback tectonics: Long-term growth of Kilauea on the south flank of Mauna Loa
Compositional and age data from offshore pillow lavas and volcaniclastic sediments, along with on-land geologic, seismic, and deformation data, provide broad perspectives on the early growth of Kilauea Volcano and the long-term geometric evolution of its rift zones. Sulfur-rich glass rinds on pillow lavas and volcaniclastic sediments derived from them document early underwater growth of a large compositionally diverse alkalic edifice. The alkalic rocks yield 40 Ar/ 39 Ar ages as old as about 275 ka; transitional-composition lavas, which mark beginning of the shield stage while most or all the edifice remained below sea level, probably first erupted after about 150 ka, and tholeiitic lavas of present-day type are probably younger than 100 ka. Breccia clasts from Papau Seamount and along the lower southwest corner of the Hilina bench are derived from subaerial Mauna Loa, requiring that Mauna Loa's flank underlies western parts of Kilauea at shallow depth. The volume of the Kilauea edifice is therefore smaller (∼10,000 km 3 ) than previous estimates (15–40,000 km 3 ); lava-thickness accumulation rates appear to have remained nearly constant during edifice growth, as effusion rates increased from ∼25×10 6 m 3 /yr at end of the alkalic stage to the present-day tholeiite rate of ∼100×10 6 m 3 /yr. Seismic and gravity data show that the deep plumbing system for Kilauea's magma supply extends nearly vertically through the oceanic crust at least to mantle depths of 30–35 km, directly below its present-day caldera. Proximity of Kilauea's caldera to the surface boundary with Mauna Loa and the presence of Mauna Loa rocks at shallow depth beneath the south flank are difficult to reconcile with a submarine origin for early Kilauea alkalic lavas, unless geometric relations between the two volcanoes have changed substantially during growth of the Kilauea shield. Seismic and ground deformation data suggest seaward spreading of the entire south flank of Hawaii Island, independently of the boundary between Kilauea and Mauna Loa, along a landward-dipping detachment fault system near the basal contact of the composite volcanic edifices with underlying oceanic crust. Current steady-state horizontal displacements increase seaward, at rates of ∼1.5 cm/yr on the lower flank of Mauna Loa and reaching 5–8 cm/yr at the Kilauea coastline. Infrequent (∼100 yr?) large earthquakes generate similar geometries, but 10 2 larger displacements per event. Present-day Kilauea is the more dynamic edifice, but prior to inception of Kilauea and during its early growth, Mauna Loa is inferred to have undergone intense volcano spreading, involving the Kaoiki–Honuapo fault system (considered a geometric analog of the Hilina system on Kilauea). Cumulative deformation of Mauna Loa's south flank during growth of Kilauea since 200–300 ka is estimated to have involved > 10 km of seaward spreading, displacing the rift zones of Kīlauea while its deep plumbing system and summit magma reservoir remained nearly fixed in space. Kilauea's rift zones, rather than migrating southward with time solely due to dike emplacement preferentially on the mobile seaward side, alternatively are interpreted to have been transported passively southward, “piggyback” style, during shield-stage growth of Kilauea as a blister on the still-mobile south flank of Mauna Loa. Such an evolution of Kilauea accounts for the arcuate geometry of the present-day rift zones, proximity of the summit magma supply to the exposed flank of Mauna Loa, initial submarine growth of the ancestral edifice, and present-day location of Mauna Loa rocks at shallow depth beneath the south flank of Kilauea.
Growth history of Kilauea inferred from volatile concentrations in submarine-collected basalts
Major-element and volatile (H 2 O, CO 2 , S) compositions of glasses from the submarine flanks of Kilauea Volcano record its growth from pre-shield into tholeiite shield-stage. Pillow lavas of mildly alkalic basalt at 2600–1900 mbsl on the upper slope of the south flank are an intermediate link between deeper alkalic volcaniclastics and the modern tholeiite shield. Lava clast glasses from the west flank of Papau Seamount are subaerial Mauna Loa-like tholeiite and mark the contact between the two volcanoes. H 2 O and CO 2 in sandstone and breccia glasses from the Hilina bench, and in alkalic to tholeiitic pillow glasses above and to the east, were measured by FTIR. Volatile saturation pressures equal sampling depths (10 MPa = 1000 m water) for south flank and Puna Ridge pillow lavas, suggesting recovery near eruption depths and/or vapor re-equilibration during down-slope flow. South flank glasses are divisible into low-pressure (CO 2 <40 ppm, H 2 O < 0.5 wt.%, S <500 ppm), moderate-pressure (CO 2 <40 ppm, H 2 O >0.5 wt.%, S 1000–1700 ppm), and high-pressure groups (CO 2 >40 ppm, S ∼1000 ppm), corresponding to eruption ≥ sea level, at moderate water depths (300–1000 m) or shallower but in disequilibrium, and in deep water (>1000 m). Saturation pressures range widely in early alkalic to strongly alkalic breccia clast and sandstone glasses, establishing that early Kīlauea's vents spanned much of Mauna Loa's submarine flank, with some vents exceeding sea level. Later south flank alkalic pillow lavas expose a sizeable submarine edifice that grew concurrent with nearby subaerial alkalic eruptions. The onset of the tholeiitic shield stage is marked by extension of eruptions eastward and into deeper water (to 5500 m) during growth of the Puna Ridge. Subaerial and shallow water eruptions from earliest Kilauea show that it is underlain shallowly by Mauna Loa, implying that Mauna Loa is larger, and Kilauea smaller, than previously recognized. Keywords
Geologic map of the central San Juan caldera cluster, southwestern Colorado
The San Juan Mountains are the largest erosional remnant of a composite volcanic field that covered much of the southern Rocky Mountains in middle Tertiary time. The San Juan field consists mainly of intermediate-composition lavas and breccias, erupted about 35-30 Ma from scattered central volcanoes (Conejos Formation) and overlain by voluminous ash-flow sheets erupted from caldera sources. In the central San Juan Mountains, eruption of at least 8,800 km 3 of dacitic-rhyolitic magma as nine major ash flow sheets (individually 150-5,000 km 3 ) was accompanied by recurrent caldera subsidence between 28.3 Ma and about 26.5 Ma. Voluminous andesitic-dacitic lavas and breccias erupted from central volcanoes prior to the ash-flow eruptions, and similar lava eruptions continued within and adjacent to the calderas during the period of more silicic explosive volcanism. Exposed calderas vary in size from 10 to 75 km in maximum dimension; the largest calderas are associated with the most voluminous eruptions.
Physical and chemical properties of submarine basaltic rocks from the submarine flanks of the Hawaiian Islands
To evaluate physical and chemical diversity in submarine basaltic rocks, approximately 280 deep submarine samples recovered by submersibles from the underwater flanks of the Hawaiian Islands were analyzed and compared. Based on observations from the submersibles and hand specimens, these samples were classified into three main occurrence types (lavas, coarse-grained volcaniclastic rocks, and fine-grained sediments), each with several subtypes. The whole-rock sulfur content and porosity in submarine basaltic rocks, recovered from depths greater than 2000 m, range from < 10 ppm and 2 vol.% to 2200 ppm and 47 vol.%, respectively. These wide variations cannot be due just to different ambient pressures at the collection depths, as inferred previously for submarine erupted lavas. The physical and chemical properties of the recovered samples, especially a combination of three whole-rock parameters ( F e-oxidation state, S ulfur content, and P orosity), are closely related to the occurrence type. The FSP triangular diagram is a valuable indicator of the source location of basaltic fragments deposited in deep submarine areas. This diagram can be applied to basaltic rocks such as clasts in debris-flow deposits, submarine-emplaced lava flows that may have crossed the shoreline, and slightly altered geological samples.
Chemical analyses of tertiary volcanic rocks, central San Juan caldera complex, southwestern Colorado
In conjunction with integrated mapping of the Oligocene central San Juan caldera cluster, southwestern Colorado (USGS I-Map 2799, in press), all modern chemical analyses of volcanic rocks for this area determined in laboratories of the U.S. Geological Survey have been re-evaluated in terms of the stratigraphic sequence as presently understood. These include approximately 700 unpublished analyses made between 1986 and 2003, as well as all USGS analyses published since 1965 when the widespread presence of regional welded ash-flow tuffs erupted from large calderas was first recognized. All the analyses are assigned unit identifiers consistent with those used for the new geologic map; quite a few of these differ from those used on sample submittal forms and in prior USGS publications.
Ups and downs on spreading flanks of ocean-island volcanoes: evidence from Mauna Loa and Kīlauea
Submarine-flank deposits of Hawaiian volcanoes are widely recognized to have formed largely by gravitationally driven volcano spreading and associated landsliding. Observations from submersibles show that prominent benches at middepths on flanks of Mauna Loa and Kilauea consist of volcaniclastic debris derived by landsliding from nearby shallow submarine and subaerial flanks of the same edifice. Massive slide breccias from the mature subaerial tholeiitic shield of Mauna Loa underlie the frontal scarp of its South Kona bench. In contrast, coarse volcaniclastic sediments derived largely from submarine-erupted preshield alkalic and transitional basalts of ancestral Kilauea underlie its Hilina bench. Both midslope benches record the same general processes of slope failure, followed by modest compression during continued volcano spreading, even though they record development during different stages of edifice growth. The dive results suggest that volcaniclastic rocks at the north end of the Kona bench, interpreted by others as distal sediments from older volcanoes that were offscraped, uplifted, and accreted to the island by far-traveled thrusts, alternatively are a largely coherent stratigraphic assemblage deposited in a basin behind the South Kona bench.
Introduction to section 2 climax-stage magmatism: Growth history of Kïlauea volcano and its instability
No abstract available.
Submarine landslides and volcanic features on Kohala and Mauna Kea volcanoes and the Hana Ridge, Hawaii
The deep submarine eastern flanks of Mauna Kea, Kohala, and Haleakala volcanoes were mapped for the first time with a multibeam bathymetric and sidescan sonar system during joint Japan-US cruises aboard the JAMSTEC vessel R/V Yokosuka in 1999. The Pololu slump off northeast Kohala is overlain by a carbonate platform in the shallow region and the deeper areas are incised by downslope oriented channels. It is cut by several faults and slump scars and appears to override an older slide located farther east, here named the Laupahoehoe slump. The structures characteristic of the Laupahoehoe slump are NW-SE oriented scarp-and-bench topographic features analogous to the Hilina slump on the mobile SE flank of Kilauea. Enclosed basins lie at 3000-5000 m, fronted by ridges on their seaward sides. The basins may result from local rotational slumps or from uplift above discontinuous thrust faults. The Laupahoehoe slump appears to be overlapped by shield margins of both Kohala and Mauna Kea and thus may have been derived from an elongate Kohala edifice. A large debris apron continues from the base of the two-slide complex, abutting the distal Hana Ridge (Haleakala east rift zone). The tip of the Hana Ridge displays a curious steepsided arcuate rift tip that resembles the classic amphitheater scarp of a landslide. Numerous flat-topped volcanic domes are distributed along the broad crest of the lower rift zone. Similar cones, though fewer in number, are also present on the Hilo Ridge, which may be the continuation of a Kohala rift zone.
Deep-sea volcaniclastic sedimentation around the southern flank of Hawaii
Most slopes of the Hilina slump are steep, but local small benches, mantled by volcaniclastic sand and fine sediments, were sampled in 1998-1999 with ROV KAIKO and DSRV SHINKAI 6500. Most surficial glass sands on the Hilina slump have compositions of subaerially erupted Kilauea lava, which fragmented and quenched as they entered the sea. Samples from the base of the Puna Ridge contain both subaerially and submarine-erupted fragments of Kilauea composition. Some glass sands from the base of Loihi contain both subaerial Kilauea and submarine Loihi compositions in the same bed. Two piston cores, collected 120 km (P6) and 250 km (P5) southeast of Hawaii, are composed of fine sediments interbedded with volcaniclastic turbidite layers. Although some volcaniclastic fragments in the fine sediments intervals are disturbed by bioturbation, nearly continuous volcanostratigraphic sequences are preserved. P6 and the upper 2.3 m of P5 are normally magnetized: the lower part of P5 is reversed and is therefore older than 0.78 Ma. At depths of 4 m below sea floor (mbsf) in P6 and 1 mbsf in P5, the dominant glass compositions change down core from Kilauea to Mauna Loa type. In the P6, an interval of abundant submarine-erupted alkalic glasses lies between 3.3 and 1.75 mbsf and may record the ancestral alkalic phase of Kilauea volcano. Magnetic susceptibility trends and glass compositions suggest that the entire P6 core correlates with only the uppermost 1.2 m of the P5, and that the average sedimentation rate at the P6 is about 5 times greater than that at P5.
Emplacement and inflation structures of submarine and subaerial pahoehoe lavas from Hawaii
Features of subaerial pahoehoe tumuli from Kilauea and Mauna Loa Volcanoes in Hawaii and subaqueous flow lobes from Loihi Seamount off Hawaii and north of Oahu Island document the controlling factors of flow-lobe formation. Studied subaerial flow-lobe tumuli consist of uplifted pahoehoe crust, formed from coalesced flow lobes. The south rift zone of Loihi has abundant conical lava mounds and terraces with flat tops. Steep flanks (>4°) of these mounds are covered with elongate pillows, but the flat tops are overlain by lobate sheets and inflated pahoehoe flows. The pahoehoe lobes have an elongate dome-like shape 2 x 0.5 m to 20 x 20 m and up to 5 m in height. Hollow pahoehoe lobes, which are rare among subaerial pahoehoe flows on coastal plains in Hawaii Island, are common on Loihi. Similar hollow lobes and a subaqueous tumulus were found on a 2200-m deep terrace offshore of Oahu Island. In contrast to corrugated pillow lobes, these subaqueous flows have smooth surfaces covered with fine streaks of thin glass flakes. We suggest that subaqueous flow lobes grow like a slowly advancing subaerial pahoehoe flow. The plastic crust stretches, thins, and ruptures at the flow front, where new lava is exposed, and subsequently is quenched by water to form thin flaky glass. The Graetz number, giving a dimensionless temperature distribution within lava, varies flow by flow but remains almost constant for subaqueous flow lobes. This means that the lateral extension of a flow lobe is cooling limited, which results in a linear correlation between the rate of lava supply and the volume of lobes. Flow lobes stop advancing due to heat loss to the environment, as brittle crust thickens and hinders the lobes from deforming freely.
Ancestral submarine growth of Kïlauea Volcano and instability of its south flank
Joint Japan-USA cruises in 1998-99 explored and sampled the previously unstudied deep offshore region south of Kilauea. Bathymetric features, dive observations, and recovered samples indicate that the 3-km-deep mid-slope bench, bounded seaward by a 2-km-high lower scarp, is underlain by massive turbidite sandstone and interbedded debris-flow breccia. Debris-flow clasts are submarineerupted (high-S) alkalic basalt, distinctive fine- to coarse-grained alkalic gabbro and nephelinite (some containing phlogopite), and subordinate transitional basalt. N o recovered clasts are similar to recent Kilauea tholeiite. Primary volcanic deposits (pillow basalt, hyaloclastite breccia) are absent. The sandstone and breccia matrix, a large fraction of the volcaniclastic apron, are mainly subaerially erupted (low-S) tholeiitic glass sand generated by shoreline processes on pre-Kilauea volcanoes. Fractures, shears, slickensided clasts, and open folds indicate widespread deformation low in the lower scarp; upward-decreasing proportions of alkalic materials define a gross stratigraphy. Alkalic high-S compositions of many basalt clasts and some sandstone glass indicate derivation from the submarine "Lo'ihi" stage of ancestral Kilauea, prior to growth of its tholeiitic shield. Slopes (3300-2800 m depth) above the mid-slope bench contain submarine-erupted (S >750 ppm) pillow lava of transitional basalt, defining the initial flank of subalkaline Kilauea. The geometry and diverse constituents of the bench and lower scarp require initial landsliding during alkalic volcanism at 200-300 ka, prior to shield growth at Kilauea and inception of Hilina faults. The active Hilina slump structures on Kilauea's south flank are in an early growth stage, thus possibly posing greater potential for future large-scale landsliding and tsunamis
Submarine alkalic through tholeiitic shield-stage development of Kïlauea volcano, Hawai’i
The submarine Hilina region exposes a succession of magma compositions spanning the juvenile "Lō‘ihi" through tholeiitic shield stages of Kïlauea volcano. Early products, preserved as glass grains and clasts in volcaniclastic rocks of the 3000 m deep Hilina bench, include nephelinite, basanite, phonotephrite, hawaiite, alkali basalt, transitional basalt, and rare alkali-poor Mauna Loa-like tholeiite. Transitional basalt pillow lavas overlie the volcaniclastic section and record an early phase of subsequent subalkaline magmatism. Rare degassed tholeiitic pillow lava and talus above the volcaniclastic section are products of subaerial shield volcanism. Major and trace element variations of clasts and pillow lavas point to a factor of 2 - 2.5 increase in degree of melting from juvenile alkalic to modern tholeiitic Kïlauea. Progressive changes in element ratios that distinguish Hawaiian shield volcanoes, without commensurate changes in elements fractionated by partial melting, also signal increased contributions from Mauna Loa-type source regions as Kïlauea matured from its juvenile alkalic to its tholeiitic shield stage. Ancestral Kïlauea basanites and nephelinites were not primitive magmas but might have evolved from plume-derived alkali picritic parents by lithospheric- level crystallization differentiation, or solidification and remelting, involving pyroxene and garnet, similar to the subcrustal differentiation origin of hawaiites [Frey et al, 1990]. Low magmatic productivity early in Kïlauea’s history sustained a poorly integrated trans-lithospheric conduit system in which magmas stalled and differentiated, producing evolved hawaiites, nephelinites, and basanites. This contrasts with shield-stage Kïlauea where high magmatic productivity flushes the conduit system and delivers primitive magmas to shallow levels.