The Geology of Crater Lake National Park, Oregon With a reconnaissance of the Cascade Range southward to Mount Shasta by Howell Williams
The Origin of the Caldera
Diller’s Collapse Theory
Diller felt convinced that Crater Lake was formed by engulfment. He argued that if 17 cubic miles of rock had been blown out by explosions, a layer of coarse fragmental debris more than a thousand feet thick would have been formed in the immediate vicinity. Even if the ejecta had been more widely scattered, there would still be an enormous accumulation near the source. He could find no trace of such a deposit. Instead, he noted, as Dutton had, that glaciated surfaces and moraines are exposed in many places beyond the caldera rim and that elsewhere they are covered by a thick blanket of pumice. He decided, further, that the pumice was erupted long before the destruction of the top of Mount Mazama.
Failing to see evidence of an explosive origin, he looked, though in vain, for positive arguments in support of the collapse hypothesis. He thought, erroneously, that the Cleetwood dacite flow was erupted during the engulfment and that part of it poured “backward” down the caldera wall. Like Dutton, he regarded the absence of arcuate fault blocks and curved fissures beyond the caldera rim as a difficulty in the way of the subsidence theory. To account for their absence, he postulated gradual, piecemeal foundering from the center outward. Moreover, being aware that engulfment demands withdrawal of magma from depth in sufficient volume to make room for the part that collapses, he pictured the formation of rifts far down the flanks of Mount Mazama and the eruption from these rifts of 17 cubic miles of lava. Nowhere was he able to find the lava for which he sought. Subsequent surveys have shown that no such voluminous flows occur anywhere in this region.
Diller was thus forced to adopt the collapse theory on negative evidence, namely, the absence of an adequate volume of fragmental debris in the vicinity of the caldera.
***previous*** — ***next***
