History of the changes in the Mount Loa craters, on Hawaii

Abstract

This series of articles, and that following, are Dana's major contribution to the understanding of Hawaii's active volcanoes, and form the basis for his later summary in book form (Dana, 1891a). The first parts summarize the history of Kilauea to 1887, compiled from numerous sources, including personal visits in 1840 and 1887. This paper is valuable, not only because it brings together the observations of many earlier workers, but also because it critically identifies the earlier observations that contribute to an understanding of volcanic processes. Dana is the first to address the mobility of islands in Halemaumau, which later became a controversial issue between the adherents of "rooted" and "floating" islands. He quotes C.S. Lyman (1851), who was the first to notice endogenous uplift of the Kilauea caldera floor, and Coan (1851), who gave the first account of doming of Halemaumau lava lake in 1848. He formalizes the concept of eruptive cycles at Kilauea and defines these as (1) filling of the caldera alternating with (2) collapse accompanied by underground flow of magma. He suspects that a submarine eruption from Kilauea was associated with the great 1868 earthquake. Dana accurately describes aa and pahoehoe, noting that one may grade into the other and that both types may be produced in a single eruption. He concludes that there is no connection between the composition and mineralogy of a flow and the type of flow surface produced. He infers that tumuli form after the original surface has cooled, from noting the direction of convexity of pahoehoe ropes (i.e., that the convexity of a ropy surface must have originally formed in the downhill direction and has been reversed by uplift at some later time.) He describes accretionary lava balls as "bomblike masses," noting that they are not bombs projected from a vent. Dana also describes fissures opening along the southwest rift zone during the collapse of 1886 (Dana, 1886), discusses the possibility that the 1832 eruption was accompanied by a collapse equal to that in 1823, and notes the possible association of the 1790 eruption with caldera collapse. He looked for deposits similar to the 1790 ash in the wall of Kilauea as an indication of earlier collapses, and found none.