The breaching of cinder cones has generally received little attention in previous works. Macdonald (1972) lists 3 means by which a cinder cone can become breached. Two other methods have been added to complete this list: 1) removal of cone material by the erupting lava after the cone has been built, 2) removal of scoria from a strombolian eruption by a simultaneously erupting lava flow preventing the side of the cone from being built, 3) shape of the vent that controls deposition of the material in the cone such that part of the cone does not get built, 4) erosion of the cone during or after the eruption, and 5) slope-failure collapse of part of the cone either during or after the eruption. This study is concerned with the first method of breaching. It should, however, be noted that both Strawberry and O'Neill Craters possess breaches which resulted from the eruption of a lava flow and collapse of the cone. The following discussion is concernedprimarily with breaching due to erupting lava after cone building has ceased.
Two questions asked during this study are: 1) why have Strawberry and O'Neill Craters been breached to the east and, 2) is there a correlation to breach directions for other cinder cones in the SFVF? Reconnaissance observations of scoria cones in the SFVF indicate that breached cones appear to have a dominant breach direction to the east. Both Strawberry and O'Neill Craters have breach directions to the east. Statistical analysis, however, documents a concentration of breach directions to the east and west.
Data collected for the determination of breach direction was compiled by examination of the MF-1956 through MF-1960 USGS Miscellaneous Investigation Series maps on the SFVF and field examination of 6 of the breached cones, two of which were studied in detail. Identification of breached cones on the geologic map was based on two criteria: 1) asymmetrical topographic morphology, primarily those with "C" shapes of the cone in plan view, and 2) association of a lava flow with these cones, especially flows that can be traced to the breach location. The azimuths of 36 breached cinder cones were located and statistically analyzed with the directional data procedure outlined by Davis (1986).The results of the analyses are presented in Table 4, Table 5 and Table 6, and Figure 39 and Figure 40.
Figure 39 shows the rose diagram, mean vector direction, resultant vector length and 95% confidence interval for the data set containing all breach azimuths. The resulting vector directions for the two subsets, S57.7E (122.3°) and N67.9W (292.1°), are approximately perpendicular to the N50E least horizontal principal stress direction (LPS) for the SFVF (Brumbaugh, 1988, oral commun.). Figure 40 shows the rose diagram, mean vector direction, resultant vector length and 95% confidence interval for azimuths east and west of longitude 111°45', an arbitrary division of the SFVF into eastern and western halves. The mean vectors are poorly constrained, but do show a general trend of west breached cones in the western SFVF and east breached cones in the eastern SFVF.
Six main factors appear to control, or influence the direction of breaching: 1) regional stress regime, 2) local and regional fault/joint systems, 3) location of the vent of the breaching lava, 4) diverting of the lava because of a substrate buttress. 5) build up and strengthening of the cone structure due to wind factors and/or the degree of welding of the deposited material, and 6) local surface topographic stress regimes.
Of the 6 breached cones that were examined in the field, none show any evidence to suggest that parallel dike swarm intrusions or major faulting were responsible for the direction of breach, as is the case for many composite cones (Siebert, 1984). Additionally, the direction of cinder cone breach is roughly perpendicular to the LPS, and parallel with the maximum horizontal compressive stress direction (MHC). This is opposite the breach/stress direction determined for composite cones (Siebert, 1984; Francis and Wells, 1988).The strong trend of breaching on SFVF cinder cones perpendicular to the LPS suggests control by the regional stress regime, but individual cones do not display recognizable field evidence for regional tectonic activity during their eruption.
The location and morphology of cinder cones in the SFVF has been correlated to the fault systems of the southern Colorado Plateau (Breed, 1964; Settle, 1979). Many cinder cones are clustered along linear trends and show elongate morphologies parallel to the fault trend. Two situations appear to permit fault control of the breach direction. The first situation is that the breach may occur along the trend of the fault (Figure 41), due to either minor motion of that fault during the eruption, weakening the cone structure, or by providing a less resistant path for the breaching lava to reach the surface. The second situation is cones whose elongate morphologies appear to be easily breached perpendicular to the elongation (Figure 41). Elongate cones also provide an easy path for breaching at the "ends" of the cone if only the ramparts along the fissure have been built.
The third control on the direction of breaching is the location of the vent from which the late stage lava erupts. Late stage lavas can erupt from a vent located at any point within the cone or long the flanks (Figure 42). It is the lava that causes the cone to become breached by removing portions of the cone during eruption. If the vent is located within the cone, the chance of the extruding lava disrupting the cone structure is increased and may result in a breach. If, however, the lava erupts along the flanks of the cone or in a portion of the cone that is relatively strong, breaching may not occur. One point to note is that the new vent locations are often oriented along the trend of the local fault/joint system.
The forth control on the breach direction is the substrate upon whichthe cone is built. The substrate can act as a buttress for the cone and force the erupting lava to take a path of less resistance (Figure 43). This appears to be best illustrated at Doney Mountain (Figure 43). Doney Mountain was built from a vent located along the Black Point Monocline fault. The cliffs immediately to the west, upon which part of the cone was built, acted as a buttress, forcing the lava flow to the east.
In wind dominated regimes, cinder cone morphology shows a marked heightincrease on the downwind side of the cone (Figure 44; Dehn, 1988; Wood, 1980). Does increased cone height produce a corresponding increase in cone strength and thus a resistance to breaching, or does the increased height produce a gravitational instability in the cone and thus a susceptibility to breaching? The initial reaction to such a question is that breaching should occur on the side of the cone which has a lower height. At Saddle Crater (Figure 44) this appears to be exactly what has happened, suggesting that increased height equates to increased cone strength. If only the height of a cinder cone is considered, however, the suggested linear relation between height and strength does not appear to hold true.
A maximum height for cinder cones is between 300 m and 400 m (Wood, 1980). This height is not exceeded even with sustained eruptions over long periods of time (Foshag and Gonzalez, 1956; Wood, 1980), suggesting that at some point during long-lived eruptions the cone becomes gravitationally unstable, and susceptible to collapse. This is believed to equate to a decrease in the strength of the cone once a certain height is achieved, followed by a point in which a portion of the cone rapidly loses strength. As the cone continues to increase in size, so does its weight, and therefore the frictional force between pyroclasts, thus increasing the strength of the cone. At some point, however, the frictional force will be exceeded by the weight of the cone, thus resulting in a decrease in the strength. It is believed that an overburden of material is needed before the cone collapses. The height at which the cone begins a decrease in strength is not known, however, the point of rapid strength loss is believed to vary between heights of 350 and 450 m. Figure 45 shows a proposed, generalized relationship between the height of a cinder cone and the strength of the cone structure. In addition to an increase in cone height, the material being deposited on the cone must also be taken into consideration. Loose, unconsolidated scoria is believed to result in a weaker cone structure than one that is dominated by or contains agglutinate or rootless flow.
The combination of the degree of welding of the deposits, controlled deposition due to wind influences and/or total height of the cone could considerably vary the strength of the cone from location to location around the cone. Height increase, alone, could also lead to an increased susceptibility to breaching.
This concept is based on the extensional and compressive stresses that develop in a free-standing body that is allowed to deform due to gravitational forces, and the resulting complimentary spreading (Figure 46; Ramberg, 1981). The local stress regime in cinder cones, therefore, can be divided in to a downward- directed gravitational compressive stress and a lateral compressive stress. The stress within the upper part of a solitary cinder cone would be extensional in a radial direction. Placement of the cone into close proximity to another topographic high, with its own lateral compressive stress, could create a local maximum horizontal compressive stress field (LMHC) in the substrate parallel to the trend of the two highs, and a local least principal stress field (LLPS) within the cone perpendicular to the LMHC. Breaching would result parallel to the LMHC due to weakening of the cone caused by minor extension along the LLPS (Figure 47). In the SFVF the direction of breaching can be found parallel to the LMHC, but only occurs on the side of the cone opposite the nearby topographic high. This model is the least supported of the local controls and is questionable as to whether its influence is great enough to control the breach direction.
The fault/joint control is believed to be the strongest influence on breach directions. Examples that appear to have originated by each of these local controls can be found within the SFVF, however, there does not appear to be a single control which dominantly influences breach directions for the entire volcanic field. This is a subject that requires further study.
The direction of breaching at Strawberry Crater appears to have been partly controlled by the surface expression of the substrate upon which the cone was erupted (control #4 - substrate buttress). The western portion of the cone at Strawberry Crater rests upon the Deadman Mesa flow. The magma body, which is believed to have intruded into the cone structure, could have encountered the Deadman Mesa flow below the west side of the cone. Given that the unconsolidated cinders and agglutinate of the cone would provide a less resistant path for the magma than the existing lava flow, the extruding lava would most likely have been forced to the north, south or east. Reasons for a breach on the east are based on the possibility that the north and south sides of the cone were stronger than the east side. This may be due to a greater abundance of agglutinate and rootless flow (control #5 - cone strength).
The reason for an east-directed breach at O'Neill Crater appears to be related to the combination of fault control (control #2 - fault/joint system) and vent location of the breaching lava (control #3 - breaching-lava vent location). The extrusion point for the lava flows is believed to be on the southeast flank of the cone. This location matches well with the local NW-SE fault trend for the southern Colorado Plateau. The reason the breach is to the east, instead of the southeast, is possibly a result of the final rafting or pushing of the large mounds to the east thus opening an eastward-directed breach.
Previous works have noted that when a cinder cone is breached, the material from the cone that is removed, is carried away by the erupting lava flow (Foshag and Gonzalez, 1956; Macdonald, 1972; and Gutmann, 1979). Holm (1987) used the rafted material to show that Sunset Crater had been formerly breached, but was rebuilt to its present symmetrical, unbreached shape. This study has taken the examination of the rafted material one step further. By determining where the rafted material is on the flow and/or where the material originated from in the cone stratigraphy, it is possible to reconstruct the breaching event in detail.
The location of the rafted mounds at Strawberry Crater and O'Neill Crater have been used in a number of different ways. At Strawberry Crater the location was used to determine: 1) the relative timing of the removal of material from the cone, and 2) that slumping of cone material onto the surface of the flow was occurring episodically. At O'Neill Crater, a non-breaching flow unit and a breaching flow unit are indicated by the presence and absence of rafted mounds on the different lava flow units.
The original stratigraphic position in the cone structure of the rafted mounds was possible to determine at Strawberry Crater. The presence of rim material on the distal portions of the flow indicate that the breaching must have first occurred high in the cone structure. The lower sections of the cone, which form a barrier across the proximal end of the flow, indicate that most of the flow had to have been extruded prior to the movement of these mounds.
Future studies of cinder cones will need to be aware of the relationship between the rafted mounds and the breaching of the cone. Mapping of the breached material may be essential in determining the eruptive history of the cinder cone. If the cone is symmetrical, the presence of rafted material on the surface of the cone may indicate that the cone was breached at some time during its eruptive history. This study has shown that the location of the rafted material can be used to reveal the sequence of events during the breaching event, especially if the original stratigraphic position of the mounds can be determined. Care needs to be taken with the interpretation of breaching events if cinders are found on the flow. Simultaneous strombolian eruption and lava extrusion can result in a breached cone and cinders on the surface of the flow (Macdonald, 1972). Conversely, a resurgence of strombolian eruption, after lava extrusion has ceased, can also deposit cinders on the surface of the flow. The third possibility is scoria deposits from other vents mantling the flow, as was the case with the Sunset Crater eruption for both Strawberry and O'Neill Craters.