Johnson, D.J., Molten core model for Hawaiian rift zones, Journal of Volcanology and Geothermal Research, 66, 27-35, 1995.
Abstract. Kilauea volcano's East Rift Zone (ERZ) is extraordinary in that abundant lateral dike intrusions and rift zone widening associated with seaward slip of the south flank over a basal fault may have allowed an extensive molten core to develop. The rift zones of Mauna Loa and the Southwest Rift Zone (SWRZ) of Kilauea do not appear to have such extensive molten cores, perhaps because of a diminished role of flank slip. A molten core may efficiently convey pressure and transport magma to distant sites along the rift zone. This concept helps to explain the long, low profile of Kilauea as viewed along the axis of the ERZ. The combined influence of a molten core and flank slip helps to explain the varied repertoire of magmatic events at Kilauea. I propose that the 1975 M 7.2 earthquake marked a fundamental change of activity at Kilauea, from shallow dike injection above the molten core and largely effusive edifice growth, to chiefly endogenous growth by deep dike intrusion below the molten core. Dike intrusion slowed during the 1983-91 Pu'u O'o eruption sequence, because inflation of the magmatic system -- the engine of intrusion and flank slip processes -- was retarded by continuous magma loss via the molten ERZ core to a low-elevation vent site.
INTRODUCTION
Kilauea and Mauna Loa are active volcanoes on the Island of Hawaii that contain magma in well defined zones. Primary magma reservoirs lie 2 to 6 km beneath each summit and are supplied from at least 50 km depth in the mantle (Eaton, 1962; Decker, 1987). Secondary magma-storage regions may locally underlie the axes of well defined rift zones, which extend many tens of kilometers outward from the central reservoirs of both volcanoes (fig. 1). Eruptions are fed directly from the central reservoir to sites at the summit of a volcano, or indirectly via the rift zones to vents along the flanks.
The basic concept of Hawaiian volcanic rift zones, particularly at the upper level, is well known; however, the details of their structure and operation continue to be a focus of research. In this paper, I consider recent work on rift zones and offer a model that attempts to blend some new ideas and yet remain simple. An important point to begin with is that the structure and dynamics may differ from one rift zone to another. Evidence supports the concept that Kilauea's East Rift zone (ERZ) has a high concentration of unsolidified melt-filled dikes between about 3 to 4 km depth. This "molten core", extending at least 40 km from the summit, transmits pressure and transports magma to the distal portions of the rift zone. I propose that dike intrusions may propagate either downward, below the Kilauea ERZ molten core, or upward, above the core. In contrast, the southwest rift zone (SWRZ) of Kilauea and the rift zones of Mauna Loa (NERZ to the northeast and SWRZ to the southwest) probably do not have well developed molten cores, although locally melt concentrations may exist. Eruptions in these areas are chiefly fed by single dikes that propagate anew from the central reservoir along the rift zone.
STRUCTURE AND OPERATION OF HAWAIIAN VOLCANIC RIFT ZONES
Present models
Rift-zone magma bodies are thought to typically form in vertical blade- like shapes (Pollard et al., 1983; Rubin and Pollard, 1987; Dieterich, 1988). Ultimately, thousands of dikes become lodged within a rift zone over the life of a volcano (Walker, 1987).
Two classes of dikes are inferred from geodetic and seismic observations of Kilauea. The most familiar and frequently analyzed class of dike is intruded at shallow crustal levels by occasional brief episodes of magma injection along propagating cracks (Klein et al., 1987). These dikes generally bottom out less than 4 km below the surface. For example, measured surface displacements for the January 1983 ERZ intrusion have been modeled to give a dike bottom at 2.4 km (Dvorak et al., 1986). SWRZ dikes intruded in September 1971 and August 1981 had modeled bottom depths of 1.5 km (Dvorak, 1990) and 3.25 km (Pollard et al., 1983), respectively. An exceptionally deep dike compared to most dikes of this class was intruded into the SWRZ in June 1982. Judging by the depths of (presumed) intrusion-related earthquakes, this dike occupies the 5 to 9 km range (Klein et al., 1987). Dvorak et al. (1986) interpreted a depth of greater than 4 km to the dike from their modeling of geodetic data.
The second class of dikes occurs at great depth. Recent analysis by Delaney et al. (1990) indicates that magma accumulates at Kilauea in a deep dike system that underlies the shallow rift-zone and summit systems. The deep system is extensive, more than 50 kilometers in length; it is apparently widening over its entire length at a slow and steady pace (Delaney et al., 1990). As this dike system occupies roughly the 3 to 9 kilometer depth range, it is referred to here as the deep dike system.
Several previous workers have also suggested a deep dike system. For example, Ryan (1988) proposed that the best sink for the dense products of fractional crystallization would be the deeper levels of the rift zones where static mechanical equilibrium with the high-density country rock is achieved. Dieterich (1988) invoked magmatic force over the entire depth of the rift zone (down to the base of the volcanic pile at a depth of 9 km) to push the unbuttressed south flank of Kilauea seaward over a deep subhorizontal fault. Dieterich's (1988) mechanical analysis implies the existence of a deep pressure source; I have found that alternate solutions with a magmatic lateral force applied only in the upper 4 kilometers or so of the edifice, assuming solely shallow dikes, yield a force balance incompatible with deep-seated slip of the south flank.
A molten core within Kilauea's ERZ was first suggested by Eaton and Murata (1960) in order to explain how a dike intrusion and surface eruption in the lower east rift zone in 1960 could have been triggered and fed from the summit reservoir more than 40 kilometers away, without also producing earthquakes along the route. In their model, summit pressure was transmitted via the molten core to initiate and drive the emplacement of a local dike that fed the eruption. Once the intrusion started, magma flowed along the rift zone core to replace magma lost by intrusion and eruption.
Gaps in seismicity between the summit and ERZ intrusion sites have been noted for a number of cases within the past three decades (Jackson et al., 1975; Swanson et al., 1976a; Swanson et al., 1976b; Klein et al., 1987). Periods of slow uplift of the middle and lower portion of the ERZ lasting several days to weeks have been observed (Fiske and Kinoshita, 1969; Epp et al., 1983; Dzurisin et al., 1984; Wolfe et al., 1987) suggesting delivery of magma through passages open to flow. Wright and Fiske (1971) argued, on the basis of the petrology of lavas erupted along the ERZ, that pockets of magma are isolated along the ERZ for periods of time before eruption. Thermal modeling by Hardee (1987) demonstrates the survivability of a molten core within the first 25 kilometers of the ERZ due to a high flux of magma through that section. However, reduced usage of the more distal portion of the ERZ, as well as the entire SWRZ, probably allows complete solidification of these passageways between events.
The seaward flank of Kilauea is observed geodetically to be moving toward the south, away from the bulk of the Island of Hawaii (Swanson et al, 1976a; Lipman et al., 1985; Delaney et al., 1990). The rift zones and summit caldera are the axis of extension, and horizontal flank motion is now thought to be accommodated by slip along a deep fault (Ando, 1979; Crosson and Endo, 1981; Eissler and Kanamori, 1987; Dieterich, 1988; Delaney et al., 1990). The zone of slip may be confined to the layer of presumably weak seafloor sediments on which the volcanic pile rests (Ando, 1979). Horizontal distances measured across Kilauea's summit and rift zones have steadily lengthened during the past two decades, a trend independent of the patterns of reservoir inflation and deflation (Dvorak et al., 1983; Lipman et al., 1985; Johnson, 1987; Delaney et al., 1990). A major movement of Kilauea's south flank occurred during the November 29, 1975, M=7.2 earthquake (Ando, 1979; Lipman et al., 1985).
Edifice widening and dike injection are intimately related. Slip on a deep subhorizontal fault not only makes room for subvertical dikes intruded within Kilauea, but maintains the stress distribution that traps dikes within the rift zone and thus sustains the rifting process (Dieterich, 1988).
Proposed "molten core" model
The model that I propose for Kilauea's ERZ includes a molten core at 3-5 kilometers depth for at least 40 kilometers of its length. This zone consists of a concentration of melt-filled dikes that remains molten and thus provides the primary path for lateral magma transport along the rift zone. Magma is "trapped" because slip of the free flank of the volcano over a fault at the base of the edifice maintains a stress distribution that constrains dike intrusion to this range of depth (Dieterich, 1988). Magma may also accumulate at this level where it is neutrally buoyant with respect to the surrounding crustal density structure (Ryan, 1987a; Rubin and Pollard, 1987; but see Dieterich, 1988). Whatever the cause, the frequency and number of magmatic intrusions centered at this depth has led to a long-lived molten zone. A schematic cross-section through Kilauea's ERZ shows the proposed arrangement of dikes and molten core (Fig. 2A).
This molten core is proposed as the immediate source of magma that is intruded both above to form shallow dikes, and below to widen the deep dike system. The path of magma within the rift zone is initially lateral through the core, and in the end subvertical into "blades" extending above and below the core. An important point of my model is that the depth of the rift-zone core corresponds to the trough in the dike-trapping compressive stress produced by south flank slip (Fig. 2B; Dieterich, 1988), but not necessarily to the level of neutral buoyancy (Ryan, 1987b). Rubin (1990, fig. 7) shows that the stress state due to density stratification is easily overwhelmed by the effect of inelastic extension. Thus, the heart of the magma-trapping zone -- where the difference between horizontal rock stress and magma pressure is greatest -- appears several kilometers deeper than the level of neutral buoyancy. Shallow dikes may not be centered at a shallow equilibrium depth as proposed by Ryan (1987b) but, more likely, are set within the upper part of a deeper dike trapping stress distribution. Deep dikes (>5 km) are believed to be associated with the lower half of the stress distribution. Thus both the deep and shallow dikes are intruded subvertically from the molten core into regions less favorable to dike intrusion, corresponding to the limbs of the dike-trapping stress curve (Fig. 2B).
The proposed model simplifies the explanation of how deep and shallow dike structures may form and operate more or less simultaneously. As noted by Delaney et al. (1990), previous work on dike emplacement at Kilauea has considered only the shallow dike system. Existing models for dike propagation imply that shallow dikes pinch out at a few kilometers depth (Pollard et al., 1983; Rubin and Pollard, 1987). For example, Rubin (1990) shows dikes bottoming out at 4 kilometers depth. This view was maintained by Delaney et al. (1990) in their arguments for a deep dike system at Kilauea. Their model contains physically separated deep and shallow dike systems.
Thus, the molten rift-zone core is perceived here as a structural element that separates shallow and deep dike systems. The molten core is approximately centered at the depth of most favorable dike-trapping stress, which is maintained over the long term by south flank slip. Fracture of brittle near-surface rock by invasion of magma from the fluid core below creates shallow dikes. Propagation or widening of deep dikes result from magma that descends from the fluid core (perhaps rich in dense products of differentiation). Although the two dike systems appear to operate on (at least partly) independent schedules and possibly with independent mechanisms, in this model they are linked by a shared magma source and a related causative stress distribution.
DISCUSSION
Kilauea's recent activity
Between eruptions, Kilauea's summit reservoir fills and pressurizes with mantle-supplied magma. A molten rift-zone core would allow pressure to increase within the rift zone as well. One of the following events is the outcome of inflation: 1) emplacement of a shallow dike above the fluid core of the rift zone (or summit), or 2) emplacement of a deep dike below the fluid core or 3) both, in concert. A dike injection event of either type will remove magma from the rift-zone core and summit reservoir, lowering the pressure of the magmatic system and, hence, diminishing the likelihood of another event until pressure is restored by mantle resupply.
Major shifts in the style of activity between deep and shallow dike intrusion from the molten core are strongly related to motion of the south flank. Dieterich (1988) demonstrates that fault slip close to a rift zone produces a stress distribution that favors trapping deep dikes. I suggest that continued south-flank creep after the 1975 M 7.2 earthquake permitted dominantly deep dike injection, below the level of the fluid core. This diversion of magma led to an abrupt drop in the frequency of shallow dike intrusions and surface effusion at Kilauea.
The SWRZ and ERZ of Kilauea served as the zone of separation between the south flank slide block and the remainder of the island during the November 1975 earthquake (Lipman et al., 1985). During the event, the rift zones locally subsided 0.5 to 0.8 m along most of their length while widening as much as 1 m (Delaney et al., 1990). These displacements suggest that an initial deep dike was emplaced in concert with the 1975 earthquake. The source proposed for the magma that filled this dike was the summit reservoir and fluid rift-zone core; the dominant direction of magma movement was downward from the fluid-filled core.
Fault slip and deep dike intrusion apparently did not entirely cease after the 1975 earthquake. The frequency of south flank earthquakes within the 5-13 km depth range increased by a factor of 8 after the 1975 event, and has been diminishing very slowly since (Klein et al., 1987). Geodetic changes measured in the eight years following the 1975 earthquake were dominated by the effect of continued fault slip combined with widening of the deep dike system (Delaney et al., 1990). One explanation for such a prolonged period of slip is that the pressure driving gradual fault movement was sustained by constant recharge of the deep dike system with magma. Occasional shallow intrusions, including one that fed the September 1977 ERZ eruption, occurred between 1975 and 1983. These events may have been triggered by increasing near-surface crustal tension above the widening deep dike and fluid core.
The Pu'u O'o eruptive sequence began in the middle ERZ in January 1983 (Wolfe et al., 1987) and has persisted for more than 8 years. Since the eruption's onset, Kilauea's summit has been deflated and no shallow dikes have intruded outside of the immediate vicinity of the middle ERZ eruption sites. Furthermore, since 1983 rates have decreased for summit widening and, perhaps, rift-zone widening (Delaney et al., 1990). This simultaneous hiatus in inflation, dike injection, and rifting may reflect the relatively low eruption site elevation, which, has permitted continuous lava discharge (via the molten rift-zone core) at low pressure.
To demonstrate the affect of vent elevation on eruption, I approximate the pressure in the deflated Kilauea summit reservoir as approximately lithostatic. With a crustal density of 2300 kg/m^3 (Kinoshita et al., 1963) and thickness of overburden of 3 km, the pressure is 68 GPa. This is sufficient to support a column of magma of density 2600 kg/m^3 (Fujii and Kushiro, 1977) to 2650 m above the reservoir, which corresponds to an elevation of 750 m above sea level. The Pu'u O'o vent rests on land with an elevation of 720 m, and the later-formed Kupaianaha vent has a base at 655 m. So, since 1983 it has not been necessary for Kilauea to pressurize and inflate in order to erupt, because the low back pressure from the column of magma reaching the vents is less than the lithostatic load at the summit. Indeed Kilauea cannot inflate because of the steady leakage of magma. This absence of inflationary pressure is likely the cause of the dearth of shallow dike intrusions, and the decrease in rate of rift and summit widening.
The 1969-1974 Mauna Ulu eruption coexisted with a relatively high rate of summit extension (Delaney et al., 1990) in addition to volcanic events at other locations (Klein et al., 1987). Perhaps the higher elevation of the Mauna Ulu eruption site (960 m) required a higher inflation level, which also triggered episodic dike intrusions and rift extension.
Long-lived rift-zone eruptions, such as the Pu'u O'o sequence, are probably aided considerably by the magma transport capability of a fluid rift-zone core. Several shield-shaped vent structures, built by pre-historic sustained eruptions, are located along the Kilauea ERZ below Mauna Ulu. One of these vents, Heiheiahulu, located at an elevation of 500 m was active c.a. 1750 (Holcomb, 1987).
Comparison with Mauna Loa and SWRZ of Kilauea
Mauna Loa's rift zones do not at present appear to have fluid cores. Almost every historic eruption of Mauna Loa started with fissure eruptions inside the summit caldera (Macdonald et al., 1983; Lockwood et al., 1985). Sometimes, summit effusion was followed by fissure eruptions that progressed as a series of outbreaks along the Southwest Rift Zone (SWRZ) or Northeast Rift Zone (NERZ). The pattern of eruption migration from summit to distal portions of the rift zone is suggestive of lateral dike propagation; the dikes originate at the summit, grow with time along the rift zone, and freeze after the eruption has ended. Each subsequent rift-zone eruption is apparently fed by a fresh dike.
Mauna Loa's rift-zone lavas have remarkably uniform compositions and lack differentiates (Rhodes, 1988). This is interpreted by Rhodes (1988) to imply that, in contrast to Kilauea's rift zones, Mauna Loa's rift zones do not contain significant pockets of stored fractionating magma. Dvorak and Okamura (1987) show that the subsidence decay rate during the 1984 Mauna Loa NERZ eruption was greater than for Kilauea ERZ eruptions at comparible distances from the summit. This observation suggests increased flow resistance in Mauna Loa's NERZ, possibly due to the lack of a fluid core.
The SWRZ of Kilauea, like Mauna Loa, may also lack a fluid core, but the evidence is ambiguous. The lateral migration of dike tips and surface vent sites directly from the summit caldera in the eruptions of 1919-1920 and 1971 (Duffield et al., 1982) is similar to Mauna Loa rift-zone events. Duffield et al. (1982) showed that in comparison with Kilauea's ERZ, Kilauea's SWRZ has had a much lower rate of historical activity, is much less voluminous; all of these factors should retard development of a fluid core. On the other hand, strongly differentiated prehistoric lavas from Kilauea's SWRZ (Wright and Fiske, 1971) imply that magma storage has occurred in the SWRZ, and the distribution of earthquakes recording a SWRZ intrusion in June 1982 shows an aseismic zone extending 10 km downrift from the summit (Klein et al., 1987), suggesting the presence of a rift-zone magma body.
A molten rift zone-core can facilitate lateral transport of magma from the summit reservoir to distant vents. A consequence of this function is a rift zone with a gradual axial slope. Figure 3 compares profiles of Mauna Loa and Kilauea, taken along the rift zone axes. Kilauea's ERZ has the lowest along-axis slope. Both of Mauna Loa's major rift zones and Kilauea's SWRZ have steeper slopes, supporting the interpretation that a molten core is not available to assist magma delivery to distant sites. A larger proportion of lavas along these rift zones are erupted close to the summit.
Rift expansion is a process that enables rift-zone intrusion to be maintained and, hence, the possibility for a rift-zone core. Without rift expansion, repeated lateral intrusions in the same direction are not permitted (Dieterich, 1988), and new dikes will be confined to the summit region, or possibly intrude at varying azimuths. Indeed, numerous eruptions from vents radial to the summit have occurred on Mauna Loa's north flank (Lockwood et al., 1988). These radial vents may record a less efficient control by rift zones on crustal stresses than exists at Kilauea, where rift-normal displacements occur more readily. Thurber and Gripp (1988) note that the slope of the depressed ocean crust below Kilauea's SWRZ is steeper than below the ERZ, which would tend to deter flank overthrusting and, in turn, rifting in that portion of the volcano. The sheer bulk of the flanks of Mauna Loa, along with Kilauea, Mauna Kea, and Hualalai volcanoes which add to its girth, may make rift expansion on Mauna Loa more difficult. Furthermore, Mauna Loa's great height, combined with deep local subsidence of the oceanic crust (Thurber and Gripp, 1988), makes the height of the active magma reservoir and dike systems above the decollement between oceanic crust and volcanic pile at Mauna Loa significantly greater than at Kilauea (Ryan, 1987). This probably decreases the likelihood that dike injection can cause slip of Mauna Loa's flanks over a basal fault.
Magma supply rate to Kilauea
Active slip of the free flank of Kilauea maintains a trapping stress within the rift zone that favors endogenous volcano growth (Dieterich, 1988). As a result few surface events occur. If the low rate of lava extrusion between 1975 and 1982 was due to diversion of magma into the deep dike system, then what was the overall magma supply rate? For a simple volume calculation, I used measured rift zone extensions (Segall and Delaney, 1988; Delaney et al., 1990) over the most active 50 km segment of Kilauea SWRZ and ERZ, and a rift zone height equal to the thickness of the volcanic pile (Thurber and Gripp, 1988). Measured extensions range up to 1.8 m and full dike heights range from 8 to 9 km. Shallow dike intrusions occurred during this time interval within most of the rift zone length considered here, so use of a full dike height is justified. The total volume of intrusion is thus approximately 500 x 10^6 m^3, and divided by the time period gives an average magma supply rate of 6.6 x 10^6 m^3/mo. Addition of the volume of extruded lava during this period, principally from the September 1977 eruption, brings the supply rate to above 7 x 10^6 m^3/mo. This value compares well with rates of lava production for periods when activity was dominantly extrusive, such as an average extrusion rate of 9 x 10^6 m6/mo for three long-lasting Kilauea eruptions (1952 Halemaumau, 1967-68 Halemaumau, and 1969 Mauna Ulu, Swanson, 1972) and an average extrusion rate of 10 x 10^6 m^3/mo for 1983-84 Pu'u O'o activity (Wolfe et al., 1987). Furthermore, the calculated value compares with an average Kilauea magma supply rate of 7.2 x 10^6 m^3/mo between 1956 and 1983 estimated by Dzurisin et al. (1984) using summit tilt data.
Caldera collapse
One proposed mechanism for major caldera collapse of Hawaiian volcanoes is by voluminous eruption from the distal or submarine portion of a rift zone, draining the shallow summit-reservoir system. A difficulty with this mechanism is that extrapolation of the relationship between vent elevation and volume of flank eruption (Epp et al., 1983) to the elevation of the distal end of Kilauea's ERZ predicts an eruption volume of only 0.7 km^3 (Holcomb et al., 1988). Actual collapse volumes are larger than this; for example, Kilauea's caldera in 1823 measured 3 km^3 in volume. A second problem is that the latest voluminous flows from the submarine portion of the ERZ appear to be older than 1000 y (Holcomb et al., 1988), while the most recent caldera collapse event that produced the Kilauea caldera of today occurred in (or in the years preceding) 1790 (Holcomb, 1987).
As an alternative mechanism, I suggest that caldera collapse follows a period of sustained south flank slip and sustained deep dike intrusion, possibly in conjunction with eruptions of decades duration along Kilauea's middle ERZ segment. The deformation field produced by combined deep dike expansion and flank slip (Delaney et al., 1990) features subsidence centered over the axis of the dike and uplift along the sides with greater uplift on the mobile flank. Delaney et al. (1990) present 1976-88 observations from a profile over Kilauea's summit that show a similar pattern. Maximum summit subsidence over 1976-88 is 1.65 m just south of the summit caldera, and between 1986 and 1988 the caldera subsided at a rate of 6 to 7 cm/year (Delaney et al., 1990). Extension of the summit region, as measured by a 21 km-long monitoring line that crosses the caldera, was 2.3 meters in 1976-89 and has continued since 1984 at an average rate of 4.6 cm/year (Delaney et al., 1990). The upper 3 km of Kilauea's summit has gradually stretched and sagged because of 1976-90 gradual flank slip and expansion of the deep dike below. Although it may take some time, continuation of this pattern could lead to caldera collapse.
Kilauea's vulnerability to caldera collapse may be increased by simultaneous sustained eruption from a low elevation ERZ site because such an eruption prevents summit inflation. Since 1982 no shallow dikes have been intruded into the shallow crust above the summit reservoir because diversion of magma to the Pu'u O'o eruption site has buffered summit pressure below the level necessary to cause shallow upward intrusion. Hence, Kilauea has been unable to heal (by shallow dike injection) extensional strain that has accumulated at shallow crustal levels.
As discussed earlier, a molten core in Kilauea's ERZ may facilitate eruption at a distant site. Also, eruption below an elevation of 750 m may continue without pressurized summit inflation required to drive it. Sustained eruption beyond 30 km from the summit reservoir, however, may be less favored, because in this region the molten core seems to be less developed. Dvorak and Okamura (1987) show that subsidence accompanying distant eruptions is hindered by increased flow resistance. The Heiheiahulu vent, located at 30 km from the summit and at an elevation of 500 m, produced circa in 1750 a shield and assemblage of flows similar to that of Mauna Ulu or the present Pu'u O'o-Kupaianaha eruption (Holcomb, 1987). Possibly this ERZ eruption, sustained for months to years, set the stage for the latest caldera collapse which, according to Holcomb (1987), began some time in the 18th century and culminated with a phreatomagmatic eruption in 1790.
ACKNOWLEDGMENTS
I appreciate helpful discussions with Carl Johnson and David Little. I thank James Ewart, Scott Rowland, Barry Voight, and Edward Wolfe for thoughtful reviews. This is School of Ocean and Earth Science and Technology contribution no. 3684.
REFERENCES
Ando, M., 1979. The Hawaii earthquake of November 29, 1975: Low dip angle faulting due to forceful injection of magma. Journal of Geophysical Research, 84:7616-7626.
Crosson, R.S., and Endo, E.T., 1981. Focal mechanisms of earthquakes related to the 29 November 1975 Kalapana, Hawaii, earthquake: The effect of structure models. Bulletin of the Seismological Society of America, 71:713-718.
Delaney, P.T., Fiske, R.S., Miklius, A., Okamura, A.T., and Sako, M.K., 1990. Deep magma body beneath the summit and rift zones of Kilauea Volcano, Hawaii. Science, 247:1311-1316.
Dieterich, J.H., 1988. Growth and persistence of Hawaiian volcanic rift zones. Journal of Geophysical Research, 93:4258-4270.
Duffield, W.A., Christiansen, R.L., Koyanagi, R.Y., and Peterson, D.W., 1982. Storage, migration, and eruption of magma at Kilauea volcano, Hawaii, 1971- 1972. Journal of Volcanology and Geothermal Research, 13:273-307.
Dvorak, J.J., 1990. Geometry of the September 1971 eruptive fissure at Kilauea volcano, Hawaii. Bulletin of Volcanology, 52:507-514.
Dvorak, J.J., and Okamura, A.T., 1987. A hydraulic model to explain variations in summit tilt rate at Kilauea and Mauna Loa Volcanoes. In: R.W. Decker, T.L. Wright, and P.H.Stauffer (Editors), Volcanism in Hawaii. U.S. Geological Survey Professional Paper 1350:1281-1296.
Dvorak, J., Okamura, A., and Dieterich, J.H., 1983. Analysis of Surface deformation data, Kilauea volcano, Hawaii, October 1966 to September 1970. Journal of Geophysical Research, 88:9295-9304.
Dvorak, J.J., Okamura, A.T., English T.T., Koyanagi, R.Y., Nakata, J.S., Sako, M.K., Tanigawa, W.T., and Yamashita, K.M., 1986. Mechanical response of the south flank of Kilauea Volcano, Hawaii, to intrusive events along the rift systems. Tectonophysics, 124:193-209.
Dzurisin, D., Koyanagi, R.Y., and English, T.T., 1984. Magma supply and storage at Kilauea volcano, Hawaii, 1956 - 1983. Journal of Volcanology and Geothermal Research, 21:177-206.
Eissler, H.K., and Kanamori, H., 1987. A single-force model for the 1975 Kalapana, Hawaii, earthquake. Journal of Geophysical Research, 92:4827-4836.
Eaton, J.P., 1962. Crustal structure and volcanism in Hawaii. In: G.A. Macdonald and H. Kuno (Editors), The crust of the Pacific Basin. American Geophysical Union, Geophysical Monograph 6:13-29.
Eaton, J.P., and Murata, K.J., 1960. How volcanoes grow. Science, 132:925-938.
Epp, D., Decker, R.W., and Okamura, A.T., 1983. Relation of summit deformation to east rift zone eruptions on Kilauea Volcano, Hawaii. Geophysical Research Letters, 10:493-496.
Fiske, R.S., and Kinoshita, W.T., 1969. Inflation of Kilauea volcano prior to its 1967-1968 eruption. Science, 165:341-349.
Fujii, T., and Kushiro, I., 1977. Density, viscosity, and compressibility of basaltic liquid at high pressures. Carnegie Institution of Washington Yearbook, 76:419-424.
Hardee, H.C., 1987. Heat and mass transport in the east-rift-zone magma conduit of Kilauea Volcano. In: R.W. Decker, T.L. Wright, and P.H. Stauffer (Editors), Volcanism in Hawaii. U.S. Geological Survey Professional Paper 1350:1471-1486.
Holcomb, R.T., 1987. Eruptive history and long-term behavior of Kilauea Volcano. In: R.W. Decker, T.L. Wright, and P.H. Stauffer (Editors), Volcanism in Hawaii. U.S. Geological Survey Professional Paper 1350:261-350.
Holcomb, R.T., Moore, J.G., Lipman, P.W., and Belderson, R.H., 1988. Voluminous submarine lava flows from Hawaiian volcanoes. Geology, 16:400-404.
Jackson, D.B., Swanson, D.A., Koyanagi, R.Y., and Wright, T.L., 1975. The August and October 1968 East Rift eruptions of Kilauea Volcano, Hawaii. U.S. Geological Survey Professional Paper 890, 33pp.
Johnson, D.J., 1987. Elastic and inelastic magma storage at Kilauea Volcano. In: R.W. Decker, T.L. Wright, and P.H. Stauffer (Editors), Volcanism in Hawaii. U.S. Geological Survey Professional Paper 1350:1297-1306.
Kinoshita, W.T., Krivoy, H.L., Mabey, D.R., and MacDonald, R.R., 1963. Gravity survey of the Island of Hawaii. U.S. Geological Survey Professional Paper 475- C:C114-C116.
Klein, F.W., Koyanagi, R.Y., Nakata, J.S., and Tanigawa, W.R., 1987. The seismicity of Kilauea's magma system. In: R.W. Decker, T.L. Wright, and P.H. Stauffer (Editors), Volcanism in Hawaii. U.S. Geological Survey Professional Paper 1350:1019-1186.
Lipman, P.W., 1980. The southwest rift zone of Mauna Loa: Implications for structural evolution of Hawaiian Volcanoes. American Journal of Science, 280- A:752-776.
Lipman, P.W., Lockwood, J.P., Okamura, R.T., Swanson, D.A., and Yamashita, K. M., 1985. Ground deformation associated with the 1975 magnitude-7.2 earthquake and resulting changes in activity of Kilauea volcano 1975-1977. U.S. Geological Survey Professional Paper 1276, 45 pp.
Lockwood, J.P., Banks, N.G., English, T.T., Greenland, L.P., Jackson, D.B., Johnson, D.J., Koyanagi, R.Y., McGee, K.A., Okamuara, A.T., and Rhodes, J.M., 1985. The 1984 eruption of Mauna Loa Volcano, Hawaii. American Geophysical Union Transactions, EOS, 66:169-171.
Lockwood, J.P., Lipman, P.W., Petersen, L.D., and Warshauer, F.R., 1988. Generalized ages of surface lava flows of Mauna Loa volcano, Hawaii. U.S. Geological Survey Miscellaneous Investigation Series Map I-1908, scale 1:250,000.
Macdonald, G.A., Abbott, A.T., and Peterson, F.L., 1983. Volcanoes in the Sea, The Geology of Hawaii. The University Press of Hawaii, Honolulu, Hawaii, 517 pp.
Pollard, D.D., Delaney, P.T., Duffield, W.A., Endo, E.T., and Okamura, A.T., 1983. Surface deformation in volcanic rift zones. Tectonophysics, 94:541-584.
Rhodes, J.M., 1988. Geochemistry of the 1984 Mauna Loa eruption: Implications for magma storage and supply. Journal of Geophysical Research, 93:4453-4466.
Rubin, A.M., 1990. A comparison of rift-zone tectonics in Iceland and Hawaii. Bulletin of Volcanology, 52:302-319.
Rubin, A.M., and Pollard, D.D., 1987. Origins of blade-like dikes in volcanic rift zones. In: R.W. Decker, T.L. Wright, and P.H. Stauffer (Editors), Volcanism in Hawaii. U.S. Geological Survey Professional Paper 1350:1149-1470.
Ryan, M.P., 1987a. Elasticity and contractancy of Hawaiian olivine tholeitte and its role in the stability and structural evolution of subcaldera magma reservoirs and rift systems. In: R.W. Decker, T.L. Wright, and P.H. Stauffer (Editors), Volcanism in Hawaii. U.S. Geological Survey Professional Paper 1350:1395-1448.
Ryan, M.P., 1987b. Neutral buoyancy and the mechanical evolution of magmatic systems. In: B.O. Mysen (Editor), Magmatic Processes: Physicochemical Principles. The Geochemical Society, University Park, Pa: 259-287.
Ryan, M.P., 1988. The mechanics and three-dimensional internal structure of active magmatic systems: Kilauea volcano, Hawaii. Journal of Geophysical Research, 93:4213-4247.
Segall, P., and Delaney, P.T., 1988. Coseismic and postseismic horizontal deformation from the 1975 M 7.2 Kalapana earthquake. EOS, Transactions of the American Geophysical Union, 69:1433.
Swanson, D.A., 1972. Magma supply rate at Kilauea Volcano. 1952-1971. Science, 175:169-170.
Swanson, D.A., Duffield, W.A., and Fiske, R.S., 1976a. Displacement of the south flank of Kilauea volcano: The result of forceful intrusion of magma into the rift zones. U.S. Geological Survey Professional Paper 963, 39 pp.
Swanson, D.A., Jackson, D.B., Koyanagi, R.Y., and Wright, T.L., 1976b. The February 1969 East Rift eruption of Kilauea Volcano, Hawaii. U.S. Geological Survey Professional Paper 891, 30pp.
Thurber, C.H., and Gripp, A.E., 1988. Flexure and seismicity beneath the south flank of Kilauea Volcano and tectonic implications. Journal of Geophysical Research, 93:4271-4278.
Walker, G.P.L., 1987. The dike complex of Koolau Volcano, Oahu: Internal structure of a Hawaiian rift zone. In: R.W. Decker, T.L. Wright, and P.H. Stauffer (Editors), volcanism in Hawaii. U.S. Geological Survey Professional Paper 1350:961-996.
Wolfe, E.W., Garcia, M.O., Jackson, D.B., Koyanagi, R.Y., Neal, C.A., and Okamura, A.T., 1987. The Pu'u O'o eruption of Kilauea Volcano, episodes 1-20, January 3, 1983, to June 8, 1984. In: R.W. Decker, T.L. Wright, and P.H. Stauffer (Editors), Volcanism in Hawaii. U.S. Geological Survey Professional Paper 1350:471-508.
Wright, T.L., and Fiske, R.S., 1971. Origin of differentiated and hybrid lavas of Kilauea volcano, Hawaii. Journal of Petrology, 12:1-65.
