Richard Harwood's Thesis

Cinder Cone Breaching Events at

Strawberry and O'Neill Craters

CHAPTER 3 - O'NEILL CRATER

Introduction

The O'Neill Crater volcanics have been divided into four units (Plate 2); cinder cone (Tocc), rafted mounds of cinder (Tocm), block lava flow (Tocf), silicic vitrophyre dome (Tocd). A fifth unit of colluvial material from the cone has also been identified as a separate unit. The volcanics appear to be underlain by Permian Kaibab Formation for the majority of the complex. Moore and Wolfe (1987) indicate that large portions of the flow and cone are underlain by older basalt flows and pyroclastic deposits. The over-all eruptive sequence at O'Neill Crater is similar to that at Strawberry Crater: (1) building of the cinder cone, (2) eruption of the lava flow, (3) breaching of the cone, (4) extrusion of the dome, and (5) post-eruption erosion and deposition.

Physical Descriptions

Cone

The general appearance of the cone at O'Neill Crater is a dull red colored, "C"-shaped cone, with minor outcrops of mantling agglutinate (Figure 24). The breach is located on the eastern side of the cone. The summit of the cone is at 2147 meters (7047 feet) altitude, rising a maximum of 173 meters (570 feet) above its base.

The unbreached exterior of the cone is approximately circular in plan view, and has an average slope of 31° on the south side and 27° on the north side. Most of the cone's exterior is composed of loose unconsolidated cinders (Figure 25). Lapilli and bombs of irregular shape are the most common, while ribbon and fusiform shapes are generally rare. Small exposures of agglutinate are found on the upper south exterior slope, and do not show a distinct morphology, other than crude bedding. The change in slope angle at the base of the cone with the surrounding topography is not sharp. Concave up morphologies for both the north and south slopes is seen. The western exterior slope consists of a large mass of cinders and agglutinate which are no longer in place. Agglutinate within this deposit is jumbled and broken, with no continuous beds being preserved.

Erosion of the exterior appears to have been most active on the north slope. The base of the north slope is a large colluvial apron. Erosion of the southern slope has been minor, with only a small apron having developed. None of the cinder and agglutinate found in the colluvium appears to be the result of slumping. Soil developement on the cone has remained relatively minor, resulting in only a thin mantling of soil.

The rim and upper interior slope consist of agglutinate beds and rootless flow. These beds form a thin, upper mantle (Figure 26). Beds range in thickness from less than 1 m to greater than 2 m. The direction and degree of dip of these beds range considerably (Figure 27), and show preservation of both inward and outward dipping beds. Most of the lower interior slope consists of loose, unconsolidated lapilli and bombs. The few outcrops of agglutinate and rootless flow that are found are not continuous and do not appear to be in place. Vegetation on the south interior slope has stabilized much of the cone. The north interior slope, however, has less ground cover and has developed a number of small scree piles on the slope.

Bombs and lapilli are generally basalt to basaltic andesite (Moore and Wolfe, 1987), with plagioclase and augite as the dominant phenocrysts. Olivine phenocrysts are common but not abundant, while hypersthene is less common. Textures of the pyroclasts range from scoriaceous to vesicular, and a porphyritic texture over-all. The texture of the matrix ranges from hyalophitic to holocrystalline. Inclusions and pyroclasts with compositions similar to the lava flow and the dome, as well as gabbroic compositions, can be found.

Rootless flows within the cone structure are uncommon, and are generally massive and structureless with occasional scoriaceous pockets. Composition of these flows is the same as the scoria. Upper and lower contacts of the flows are sharp to gradational with loosely to highly welded scoria. Individual rootless flows are not continuous across the outcrop, showing a lens-like profile.

Rafted Mounds

Rafted mounds of pyroclasts on the lava flow came from the breached portion of the cone. Mounds range in size from individual blocks to mounds greater than 90 m (300 feet) high. In many areas of the flow, large, thin, unmapped sheet-like cinder deposits from the cone can be found. Material found in the mounds and the sheet-like deposits is indistinguishable from the material in the cone. Agglutinate beds and rootless flows are rarely continuous, however, and are generally broken into large blocks. The largest mounds are found immediately to the east of the cone. The remainder of the rafted mounds are restricted in location to the central and northeastern portion of the flow (Plate 2). No mounds are found on the southern or western section of the lava flow (Plate 2).

Lava Flow

The lava flow at O'Neill Crater is classified as a basaltic andesite block flow (Moore et al., 1974; Moore and Wolfe, 1987). Hand sample examination shows that the rocks range from vesicular to highly vesicular, with a porphyritic texture and an aphanitic to glassy matrix. Phenocrysts of plagioclase, pyroxene and olivine are common. Plagioclase is abundant and shows both sieve and solid crystal textures. Olivine is common, but not abundant. Quartz crystals and gabbroic inclusions are rare; calcite and tridymite can be found lining some of the vesicles. Thin section examination shows a porphyritic texture, with a hypocrystalline to hyalophitic matrix. Plagioclase, augite and hypersthene are present as phenocrysts and groundmass minerals. Plagioclase has sieve, embayed and solid crystal textures. Magnetite is are as a phenocryst. No phenocrysts of olivine were seen in thin section; it was only found in the groundmass. The glass matrix is brown to black in color. Vesicles are common, showing irregular outlines.

The flow consists of polygonal blocks, which are equidimensional and generally angular. Weathering of the blocks throughout much of the flow has rounded many of the sharp edges on the blocks. Polygonal pavement is rare, covering both large and small areas (Figure 28). Most of the flow surface is covered by a fairly thick development of soil and Sunset Crater ash and lapilli deposits. Because of the soil, outcrops are generally small on the surface of the flow. The flow edges provide the best exposures (Figure 29). An escarpment of blocks, up to 24 m (80 feet) high, is found on the interior of the flow, roughly parallel to the southern edge (Figure 30). This escarpment roughly defines the southern extent of the rafted mounds on the flow.

The flow covers an area of 7.9 km2 (3.05 mi2) and has a maximum flow length of 3.4 km (2.1 mi). Thickness of the flow varies considerably. The western flow front is less than 3 m (10 feet) high (Figure 31). The southern and eastern edges vary from 12 m to 18m (40 to 60 feet) (Figure 32). Volume of the flow is estimated at .086 km3 (.021 mi3)

Dome

Unlike the dacite plug at Strawberry Crater, the dacite vitrophyre (Moore and Wolfe, 1987) at O'Neill Crater can be classified as either a small upheaved dome or small pelean dome. The dacite dome is located in two main areas, with the large dome occupying an area slightly off center of the cinder cone, in the breach (Plate 2). The largest outcrop is a maximum of 21 m (70 feet) high, as measured from the lowest point on the dome. Outcrops are blocky, often showing large unbroken surfaces (Figure 33). Samples are generally friable, tending to crumble easily. Inclusions of basaltic andesite, lava flow-like material, gabbroic rocks and large amphibole crystals (3 cm) can be found.

Few distinct structures are readily identifiable. Small fan structures (Figure 34), and small fins and spines (Figure 34) are seen. The dome, however, lacks an appreciable crumble breccia.

Within the large fan of cinders and agglutinate on the western exterior slope of the cone, two small outcrops of dacite vitrophyre are found (Plate 2 ; Figure 35). Both are composed of rounded to angular blocks, which appear to be in place, and not a part of the fan deposit.

In hand sample, the dacite vitrophyre has a porphyritic texture, with a black to light gray, aphanitic to glassy matrix. Sieve and solid texture phenocrysts of plagioclase dominate the assemblage, reaching sizes of 1 cm. Pyroxene phenocrysts are common, while olivine and quartz are rare. Vesicles are generally less than 2 mm. Thin section examination shows a hyalopilitic to hyalophitic texture. Plagioclase crystals dominate the assemblage, having sieve, embayed and solid crystal textures, and are weakly zoned. Hypersthene and augite are present as phenocrysts and groundmass crystals. The matrix is dominated by black to brown glass.

Moore and Wolfe (1987) indicate airfall dacite fragments were associated with the eruption of the dacite dome. These deposits were not mapped by Moore and Wolfe (1987); sample locations show them as coming from a cinder pit located to the northwest of the cone (Plate 2). Association of the dacite with the O'Neill Crater eruption is based on chemical and petrographic similarity. These fragments are angular, glassy, coarse lapilli and blocks (up to 1 m in diameter) and show a phenocryst assemblage similar to the dacite vitrophyre in the dome. Some of the blocks are coated with what appears to be a mantle more mafic than the dacite (Figure 36). Hand sample and thin section examination of the mafic mantle shows a distinct petrographic assemblage different from the dacite. Olivine and augite phenocrysts dominate the assemblage with plagioclase common, but not abundant. The matrix is hyalopilitic to hyalophitic, and vesicular to scoriaceous. Although the mining operations disturbed the fragments, one outcrop indicates that the fragments were deposited above the basaltic ash and lapilli deposits of the cone (Figure 37).

Eruptive History

Cone Building

The initial eruption at O'Neill Crater was probably from a pipe-like conduit. No deposits or deflection of the lava flow around buried deposits are found to support a fissure eruption. The symmetrical morphology of the unbreached portion of the cinder cone indicates, based on Breed's (1964) work, that the vent shape for the cone building phase was circular.

The morphology of the cone and the cinders with which it is composed indicate that the eruption style at O'Neill Crater was strombolian. For the most part, the eruption appears to have been fairly consistent through time, with only slight magmatic variations occurring toward the end of the cone building period. Lower stratigraphic sections in the cone are dominated by unconsolidated lapilli and bombs, suggesting a typical strombolian style of eruption. The rim deposits show minor amounts of agglutinate and rootless flow, indicating a change in the physical characteristics of the eruption. As indicated at the Paricutin eruption (Foshag and Gonzalaz, 1956), the deposition of rootless flow is a strong indicator that the crater may have been occupied by a lava lake. Cinder cone eruption models which indicate late stage agglutinate deposition (McGetchin et al., 1974; Gutmann, 1979; and Crowe et al., 1981) appear to work well for O'Neill Crater. The main difference from the models is that the agglutinate deposits at O'Neill Crater appear to be more abundant than just a final mantling rim, suggesting that the final stage began earlier than normal.

Variations in the eruption through time, other than the final stage of cone building, have not been preserved. With the lower sections of the cone being composed of unconsolidated cinders, no changes in the strength of the eruption during cone building are indicated.

Lava Flow - Breaching of the Cone

At O'Neill Crater the locations of the rafted mounds are restricted to the central and eastern portions of the flow, while the southern and western areas are totally free of any mounds. This observation can be explained by the following two models: (1) that breaching occurred during the initial extrusion of the lava flow, with the breach becoming stabilized shortly after the breaching event, and (2) that there are two flow units, with the first flow unit as a non-breaching flow, and the second flow unit as the breaching flow. In both models, lava extrusion is believed to have begun at or near the base of the cone. There is no evidence to support the conclusion that the magma intruded into the upper cone structure or overtopped the rim.

1. Initial breach model.

This model starts with the hypothesis that the initiation of lava extrusion resulted in the breaching of the cone. Initial breaching of the cone would have resulted from material from the cone slumping onto the surface of the flow. In order to explain the distribution of the rafted mounds, the breaching process must have ceased at some point during the lava extrusion due to stabilization of the walls of the breach.

Two main problems occur with this model. The first problem is moving the southern portion of the flow to its present position. It is possible for this lava to have flowed around the western part of the area containing the breached material. The fault with this proposal is that the portion of the flow containing the rafted material appears to overlie the southern portion of the flow. The second problem with the model is that the apex area of the flow is covered with unconsolidated cinders and small mounds of agglutinate, indicating that the rafting of cone material was still occurring at the termination of lava extrusion.

2. Two flow unit model.

The model that appears to explain the lack of rafted mounds on the southern and western portions of the flow as well as the mid-flow escarpment is one in which the lava extrusion is divided into two flow units. The first flow is a non-breaching flow which is believed to have extruded to the west and south, following a paleo-Rio de Flag/San Francisco Wash drainage. The topographic expression of the first flow unit along the northern boundary of map sections 3, 4 and 5 (Figure 38) appears arched in a north to south cross-section and is linear from east to west, suggesting that the flow was following the previous topographic low of a paleo-Rio de Flag/San Francisco Wash drainage. At the western edge of the flow, the Rio de Flag is deflected around the flow and closely follows the southern edge of the flow. At the eastern extent of the first lava unit, the San Francisco Wash, which merges with the Rio de Flag along the southern edge of the flow, appears to resume its original path, approximately at the center of the first flow (Figure 38; Plate 2).

After the first flow unit had been extruded, the cone was breached by the second flow unit. This flow unit was extruded to the southeast, with the southern edge of the unit having been stopped from further movement by the first flow unit. It is not possible to determine whether or not there was a hiatus in the lava extrusion, which was followed by the breaching of the cone. It is possible that lava extrusion was continuous. The flow is divided into two units based on the morphology of the two sections of the flow, i.e. the presence of rafted mound versus a lack of rafted material, and the presence of a second flow front in the middle of the flow.

The breaching event.

The initial extrusion of lava, which did not breach the cone, appears to have had a vent located at or near the southeast base of the cone. The reason why this flow unit did not disrupt the cone is not known. Macdonald's (1972) discussion on lava extrusion events associated with cinder cones seems to indicate that the eruption of lava from a vent within the cone will disrupt the cone structure. Theoretical models of cinder cone growth indicate that lava extrusion should postdate the eruption of the cone. This raises the question of why there are few clearly documented cases of lava extrusion post-dating the cone building, which did not disrupt the cone structure. Eye-witness observations of recent cinder cones with associated lava flows show the cones as having been disrupted by the erupting lava flow (Foshag and Gonzalez, 1956; Thorrenssen, 1973). Continued pyroclastic eruption at some of these cones has rebuilt the cone resulting in a symmetrical morphology. The point is that lava extrusion post-dating the cone building should damage or breach the cone. A survey of SFVF cinder cones, however, appears to indicate that symmetrical cones with associated lava flows are the norm. One way to explain this apparent contradiction is to have the lava flow occur first, followed by cone building. This sequence of events is suggested by Ulrich (1987) for S P Crater.

Three possibilities explain why the first flow was a non-breaching flow and why the second flow breached the cone. The first is that the vent location for the first flow was positioned such that it was not in direct contact with the cone and therefore was unable to breach the cone, while the second flow unit's vent was located within the cone structure.

The second possibility is that the first flow unit was erupted prior to the building of the cone. The second flow unit would then have been erupted from a vent within the cone structure, after the eruption of the cone.

The third possibility is that both flows were using the same conduit that was used during cone building. The first flow unit could have burrowed along the base of the cone, as described by Macdonald (1972), emerging from the southeast flank. This could have caused minor amounts of deformation to the cone structure, but not enough to cause breaching. The second flow unit, following a similar pattern as the first unit, could have increased the structural deformation to the point where the cone became unstable.

No matter which model is used, the second flow is believed to have breached the cone. Slumping of cinders and agglutinate onto the flow surface, during eruption, started the breaching process. The large mounds of rafted material immediately to the east of the cone suggest that during the late stages of the flow, extrusion the eastern portion of the cone was carried or pushed away as a large mass. The opening that was created allowed the final portions of the lava to flow south through this gap. The lava extruded after the gap was created has the appearance of cascading out of the vent area (Plate 2). The presence of large amounts of unconsolidated cinders and small mounds in the breach of the cone, at the apex of the flow, indicates that slumping of material onto the flow surface was occurring at the termination of lava extrusion.

Dome Extrusion

The final eruptive event at O'Neill Crater was the extrusion of the dacite vitrophyre dome. The main mass of the dome, located in the breach, shows indications of being either a small pelean dome or a small upheaved dome, as defined by Williams and McBirney (1979). Evidence for the pelean style eruption is seen in the small fans, fins and spires that are found. The lack of a crumble breccia, however, suggests that either the dome was to small to have developed a breccia, or that it is an upheaved dome. According to Williams and McBirney (1979), an upheaved dome would possess a larger conduit than that found for a pelean dome. The small scale of the dome at O'Neill Crater makes this criteria difficult to judge. The presence of a number of small fan structures instead of a single, large fan structure for the entire dome, suggests that there might have been a large vent through which the magma was erupting. This is also supported by the wide outcrop of the dome. The small outcrop on the east side is thought to be a part of the main mass, but separated by surficial deposits (Plate 2).

The two small masses of dacite to the west of the cone pose something of a problem. Both are located in slump material, yet neither outcrop appears to be part of the slump. If these outcrops are part of the slump material the question arises as to where this material came from. There is no evidence of large outcrops of dacite vitrophyre in the cone structure, therefore they are here classified a plugs originating from separate vents.

The unconsolidated fan to the west of the cone appears to be the result of a large slump. There is no evidence to suggest that the material was deposited as a result of stream erosion of the cone. Moore and Wolfe (1987) show the mass as an andesitic lava flow similar to the main flow. Close examination of the fan, however, reveals a lack of any material that could be classified as a lava flow. The over-all shape of the fan suggests a small flow lobe. The problem with this is the absence of any flow outcrop. The two plugs appear to have been intruded into the slump, indicating that this material had to have been deposited prior to the eruption of the plugs. When this slump occurred is not possible to determine other than it predates the eruption of the two plugs. It is possible that the slump was a result of breaching or could have been triggered by tremors from the erupting dome.

The association of airfall dacite fragments with the eruption of the dome (Moore and Wolfe, 1987) indicates that some explosive activity was accompanying the magma extrusion. Williams and McBirney (1979) indicate that pelean eruptions, which result in airfall deposits, are most violent at the onset of the eruption. This could explain how the block-sized fragments were deposited at their present distance of approximately 1 km (.6 miles) from the dome. Examination of the area around the cinder pit and on the volcanics at O'Neill Crater reveals a distinct lack of dacite fragments. This is either due to a very restricted deposition of the material or more probably a result of burial by Sunset Crater ash and lapilli and the development of soil. The basaltic, scoriaceous mantle on some of the dacite fragments consists of phenocrysts of olivine, augite and plagioclase in proportions distinctly different from the basaltic andesite of O'Neill Crater. This indicates that the basaltic mantle is a different magma, apparently one which did not erupt at O'Neill Crater, but that existed in the magma chamber. These relations can be interpreted as supporting Moore and Wolfe's (1987) suggestion that the dacite may have existed as a segregated liquid contemporaneously with basalt in the magma chamber prior to eruption. Eruption of the dacite fragments appears to indicate that these pyroclasts were, at some time during the eruption, in contact with the basaltic liquid, thus becoming coated. The basaltic magma may have entered the magma chamber after the eruption of the basaltic andesite. This introduction of the basaltic liquid may have increased the pressure in the magma chamber, resulting in the eruption of the dacite.

Post-eruption events

The major post-eruption event at O'Neill Crater has been erosion and soil development on the volcanics. On the north side of the cone is a large colluvial apron. The development of this deposit appears to be strongly related to the greater amount of precipitation that O'Neill Crater receives as compared to Strawberry Crater. The apron's position on the north side is probably explained by the fact that when the snow melts it is protected from evaporation by the sun and thus tends to runoff, washing material off the cone. Soil development on the rest of the cone due to the precipitation has allowed for a greater support of vegetation. Most of the flow is covered with a moderate to thick development of soil, obscuring most of the outcrops. Despite the soil development, the flows and cone are undissected. Individual blocks indicate the beginnings of weathering with edges tending to be rounded.

Other than continued erosion, the final post-eruption event was the deposition of ash and lapilli from the eruption of Sunset Crater. Deposits from this eruption do not appear to be as prevalent as those at Strawberry Crater. This is either due to less material being deposited, or the development of soil obscuring most of these deposits.