The Dacite Flow of Llao Rock

The thickest sheet of lava on the walls of Crater Lake forms the imposing cliff known as Llao Rock. In striking contrast with the underlying flows of andesite, which are between 20 and 100 feet thick, the dacite of Llao Rock reaches a maximum thickness of no less than 1200 feet. It was erupted into a hummocky glacial valley, approximately U-shaped in cross section, about 1/2 mile wide and from 500 to 600 feet deep. Not only did the dacite fill this valley, but it accumulated as a broad, domical pile above it and spread across the valley rims. Naturally, where the lava overflowed the valley the thickness was much less than over the center. On the west bank, the thickness is at most 400 feet; on the opposite bank, it is only about half as much. The total width of the flow is slightly more than 1 1/4 miles, and its volume approximates 1/4 cubic mile.

Though the flow is exceptionally thick, it spread little more than a mile beyond the rim of the caldera. This was partly because the slope of the glacial valley was gentle, but mainly because of the highly viscous nature of the flow.

Source of the flow. Beneath the lowest part of the Llao dacite on the caldera wall are two thin dikes. Neither dike can be seen to connect with the lava, and it seems unlikely that, being only 5 and 7 feet thick respectively, they could have served to erupt 1/4 cubic mile of viscous dacite. For this reason, some have suggested that the lava was erupted from a vent higher up the slopes of Mount Mazama. This theory, however, is also untenable.

Only by scrambling up the cliffs to the base of Llao Rock can the true relations be observed. There is then no doubt that the buttress which projects lakeward at the bottom of the cliffs represents the actual filling of the conduit. In other words, the Llao lava escaped from a vertical-sided vent on the bottom of a glacial valley, several thousand feet below the former summit of Mount Mazama. The photograph (plate 12, figure I) and the section (figure 10) show the projecting buttress and feeder in relation to the main body of the flow.

The flow. The eruption of the Llao dacite was preceded by explosions of dacite pumice. The products of these explosions may be examined on the east side of the glacial valley, where they vary in thickness between 20 and 50 feet. At the base, they are incoherent or but weakly compacted, and consist of coarse lump pumice crowded with angular blocks of andesite. Some of the foreign blocks measure a yard across, but most are less than 6 inches in diameter. Their number is such as to indicate the blasting of a large vent. In its upper part, the lump pumice becomes increasingly compact and welded. Close to the base of the overlying lava it develops such a strong, streaky banding and is so dense that it is difficult to say where the flow begins and the welded pumice ends. Presumably it was the heat and weight of the lava above which caused collapse and welding of the topmost pumice.

If the portion of the Llao lava lost during the formation of the caldera were restored, a large dome would be seen above the vent. The internal structure of this dome would probably be fan-shaped, the flow planes steepening upward and inward. Clearly, the lava can never have extended far toward the south, for its movement in that direction was opposed by the slope of the main cone of Mazama (section, figure 10).

The first lava to escape moved slowly the attitude of down the glacial valley. Perhaps at an s to the dip at early stage a dome formed above the ion shows the orifice, and as this continued to grow by flow before andesites and by expansion from within, the concentric flow bands were disrupted in the central parts and unfolded like the petals of an opening bud. Shortly, the lava filled the old glacial valley and spread onto the surrounding flats.

With a little care, one may examine the flow on the west side. At the base, the dacite is a jet-black glass, streaked in shades of pink and gray. In many places it is highly autobrecciated, and the shattered bottom is rucked up, forming many small caves. Within and just above the blocky basal layer, the flow planes are contorted into flat, recumbent folds, suggestive of viscous drag. Higher up, the contortion rapidly disappears, and the flow planes become smooth and regular. At first they dip into the caldera wall at low angles, conforming to the valley floor; farther up the cliffs, they become horizontal and then gradually the dip changes in direction, becoming increasingly steep and finally, near the surface, becoming vertical.

Beyond the caldera rim, the lava is almost entirely concealed by a blanket of pumice. Clearly, however, the surface of the flow has not been greatly eroded and preserves much of its original form. Only the low fringes of the lava were later covered by ice. The crust must once have bristled with black, glassy spines and turrets and been littered with angular blocks of obsidian. The attitudes of the flow planes (figure 10) resemble those seen in the Watchman, Cleetwood, Grouse Hill, and Redcloud lavas.

When the new highway was made to Diamond Lake, a cave was discovered at the edge of the Llao flow. Allen¹ records that it measures 20 feet long and between 10 and 15 feet in diameter, and that many tributary channels a foot or so across ramify downward from the floor. On the roof of the cave he found abundant specular hematite, some of it in beautiful dendritic forms.

Whereas the basal layer and the crust of the Llao dacite consist chiefly of dense, black obsidian, the bulk of the flow consists of paler, streaky glass which is extremely vesicular and locally almost pumiceous. The rapid alternation of light and dark, that is, of more and less vesicular layers, seems to reflect an unequal concentration of gas within the rising magma.

Reference should be made, in conclusion, to the glacial deposits beneath the lava on the caldera walls. Along the east side of the old valley down which the lava flowed, glacial deposits are absent. The dacite there rests on the coarse lump pumice blown out during the blasting of the vent. Under the west "wing," on the contrary, the lava rests on glacial moraines and fluvioglacial sands (see plate 12, figure 2). Close to the edge of the flow, these deposits approximate 100 feet in thickness and contain much reassorted pumice. Toward the vent they thin, and finally, a short distance down the side of the old valley, they disappear. Striated boulders, up to 18 inches across, occur just below the lava, and occasionally one may find erratics partly embedded in the obsidian, their crusts reddened by heat and escaping gases.

<< Previous | Table of Contents | Next >>