The Eruptive History of Mount St. Helens
by Donald R. Mullineaux and Dwight R. Crandell
ABSTRACT
The eruptive history of Mount St. Helens began about 40,000 years ago with dacitic volcanism, which continued intermittently until about 2,500 yr ago. This activity included numerous explosive eruptions over periods of hundreds to thousands of yr, which were separated by apparent dormant intervals ranging in length from a few hundred to about 15,000 yr. The range of rock types erupted by the volcano changed about 2,500 yr ago, and since then, Mount St. Helens repeatedly has produced lava flows of andesite, and on at least two occasions, basalt. Other eruptions during the last 2,500 yr produced dacite and andesite pyroclastic flows and lahars, and dacite, andesite, and basalt airfall tephra. Lithologic successions of the last 2,500 yr include two sequences of andesite-dacite-basalt during the Castle Creek period, and dacite-andesite-dacite during both the Kalama and Goat Rocks periods. Major dormant intervals of the last 2,500 yr range in length from about 2 to 7 centuries.
During most eruptive periods, pyroclastic flows and lahars built fans of fragmental material around the base of the volcano and partly filled valleys leading away from Mount St. Helens. Most pyroclastic flows terminated with 20 km of the volcano, but lahars extended down some valleys at least as far as 75 km. Fans of lahars and pyroclastic flows on the north side of the volcano dammed the North Fork Toutle River to form the basin of an ancestral Spirit Lake between 3,300 and 4,000 yr ago during the Smith Creek eruptive period, and again during the following Pine Creek eruptive period.
ERUPTIVE PERIODS AT MOUNT ST. HELENS
The eruptive history of Mount St. Helens is subdivided here into nine named eruptive "periods," which are clusters of eruptions distinguished by close association in time, by similarity of rock types, or both. The term "eruptive period" is used in an informal and largely arbitrary sense to divide the volcano's history into convenient units for the purpose of discussion. The periods are as much as several thousand years in duration, and include what may have been a single group of eruptions as well as extended episodes of volcanism, during which there were tens or possibly hundreds of eruptions. Eruptive periods are separated by apparently dormant intervals, which are inferred chiefly from buried soils and absence of eruptive deposits. However, some dormant intervals may span times of minor activity that did not produce deposits which can now be recognized. Fine-grained, air-laid volcanic detritus was deposited during some dormant intervals, but these deposits are not known to have originated directly from eruptions; they might be material reworked from the flanks of the volcano.
The stratigraphic record of eruptive activity during the last 13,000 yr is believed to be reasonably complete. Parts of the older record, however, apparently are missing because of glacial and stream erosion during the last major glaciation (the late Pleistocene Fraser Glaciation) of the region.
APE CANYON ERUPTIVE PERIOD
The first stratigraphic evidence of the existence of Mount St. Helens consists of voluminous dacitic deposits of slightly vesicular to pumiceous air-fall tephra and pyroclastic flows, and at least one pumice-bearing lahar. These deposits overlie extensively weathered glacial drift formed during the next-to-last alpine glaciation of the Cascade Range. The volcanic deposits were formed during at least four episodes, separated by intervals during which very weak soils developed. The entire eruptive period may have extended over a time span as long as 5,000 yr. One pumiceous tephra deposit produced during the period probably had a volume as great as that of any subsequent tephra erupted at Mount St. Helens.
The Ape Canyon eruptive period was followed by a dormant interval which may have lasted from about 35,000 to 20,000 yr ago. Most of this 15,000-yr interval coincided with climates which, at times, were evidently somewhat cooler than those of the present (Alley, 1979, p. 233).
COUGAR ERUPTIVE PERIOD
The second eruptive period probably began about 20,000 yr ago, and was characterized by the eruption of small volumes of pumiceous dacite tephra; it also produced lahars, pyroclastic flows of pumiceous and lithic dacite, a few lava flows of dacite or high-silica andesite (C.A. Hopson, written commun., 1974), and perhaps one or more dacite domes. Several different eruptive episodes can be identified during the period. At least one pumiceous pyroclastic flow moved southward to at least 16 km from the center of the present volcano about 20,350 yr ago (Hyde, 1975, p. B11-B13). Two sequences of air-fall tephra that followed (sets M and K) are separated by a two-part deposit of fine air-laid sediment that locally is a meter or more thick, and that contains at least one weakly developed soil. After another quiet interval during which there was a small amount of soil development, at least two more pyroclastic flows moved south and southeast from the volcano between about 19,000 and 18,000 yr ago. The Cougar eruptive period occurred during the Frasier Glaciation when alpine glaciers in the Cascade Range were at or near their maximum extents, and the products of eruptions generally are poorly preserved.
One lahar that apparently occurred early in the Cougar period is of special interest because of some similarities to the debris avalanche of May 18, 1980, that swept down the North Fork Toutle Valley. The lahar of Cougar age consists of an unsorted and unstratified mixture of gray dacite fragments in a compact matrix of silt and sand as much as 20 m thick. Locally, it contains discrete texturally similar masses of red dacite many meters across. The iron-magnesium mineral content of rocks in the lahar is similar to that of the Ape Canyon period, suggesting that the lahar might have been derived from older parts of the volcano. The lahar was recognized in the Kalama River drainage 8 km southwest of the center of the modern volcano, and on both walls of the Lewis River valley near Swift dam (Hyde, 1975, p. B9-B11). It has not been recognized elsewhere; thus, little is known of its original extent. Its local thickness and heterolithologic character suggest that the lahar might have originated in a large slope failure on the south side of the Mount St. Helens of early Cougar time.
There is no stratigraphic record of volcanism at Mount St. Helens between about 18,000 and 13,000 yr ago.
SWIFT CREEK ERUPTIVE PERIOD
The third eruptive period was characterized by repeated explosive eruptions that initially produced many pyroclastic flows as well as pumiceous air-fall tephra deposits, some of which had large volumes and extended at least as far east as central Washington. These eruptions of dacite pumice were followed by many lithic pyroclastic flows, which are believed to have been derived from domes; at least one of these pyroclastic flows reached a point 21 km from the center of the present volcano. The pyroclastic flows were followed, in turn, by another series of explosive eruptions that produced the voluminous tephra set J. One coarse pumice layer of set J extends west-southwest from Mount St. Helens, and is as much as 20 cm thick as far as 20 km from the volcano. The layer represents the only coarse and thick pumice known to have been carried principally in a westerly direction. The sequence of explosive eruptions that formed set J apparently ended the Swift Creek eruptive period sometime before 8,000 yr ago, and was followed by a quiet period of at least 4,000 yr.
SMITH CREEK ERUPTIVE PERIOD
Multiple explosive eruptions of the Smith Creek eruptive period, which began about 4,000 yr ago, initiated at least 700 yr of intermittent and at times voluminous eruptive activity. Three coarse pumice layers at the base of tephra set Y are overlain by layers of denser, somewhat vesicular tephra. Deposition of these units was followed by an interval during which a soil began to develop on the tephra. The next eruption of the period produced the most voluminous and widespread tephra deposit of the last 4,000 yr; it is one of the largest, if not the largest, in the history of the volcano, and has an estimated volume of at least 3 km. The resulting pumice layer, Yn, has been found nearly 900 km to the north-northeast in Canada (Westgate and others, 1970, p. 184). The formation of this layer was followed shortly by another voluminous eruption of tephra, which resulted in layer Ye (Mullineaux and others, 1975, p. 331), then by a pumiceous pyroclastic flow and a coarse lithic pyroclastic flow. The lithic pyroclastic flow was accompanied by clouds of ash that spread at least a kilometer beyond the sides of the flow and as much as 2 km beyond its front. Many smaller eruptions of lithic and moderately vesicular ash and lapilli followed, perhaps within a few years or tens of years.
Lahars and pyroclastic flows of Smith Creek age formed a fan north of the volcano, and lahars extended down the North Fork Toutle River at least as far as 50 km downvalley from Spirit Lake. An ancestor of the lake probably came into existence at this time, dammed in the North Fork valley by the fan of lahars and pyroclastic-flow deposits. It is not known if the lake ever existed before Smith Creek time.
A dormant interval of apparently no more than a few hundred years followed the Smith Creek eruptive period.
PINE CREEK ERUPTIVE PERIOD
Although only a short time elapsed between the Smith Creek and Pine Creek periods, eruptive products of Pine Creek age contain an iron-magnesium phenocryst assemblage that is distinctly different from those of Smith Creek age. During the Pine Creek eruptive period, large pumiceous and lithic pyroclastic flows moved away from the volcano in nearly all directions. The lithic pyroclastic flows, some of which extended as far as 18 km from the present center of the volcano, are believed to have been derived from dactic domes. Eruptions of dactic airfall tephra were of small volume, but at least four formed recognizable layers as far away as Mount Rainier (Mullineaux, 1974, p. 36).
During this time, lahars and fluvial deposits aggraded the valley floors of both the North and South Fork Toutle River, and created the basin of Silver Lake 50 km west-northwest of the volcano by locking a tributary valley (Mullineaux and Crandell, 1962). Similar deposits also formed a contiguous fill across the floor of the Cowlitz River valley near Castle Rock that was about 6 m above present river level; this fill probably extended 209 km farther to the mouth of the Cowlitz River. Lahars and fluvial deposits formed a similar fill in the Lewis River valley which, near Woodland, was about 7.5 m higher than the present flood plain (Crandell and Mullineaux, 1973, p. A17-A18).
The eruptions of Pine Creek time extended over a period of about 500 yr. No single eruption of very large volume has been recognized from deposits of Pine Creek age, and the period seems to have been characterized by many tens of eruptions of small to moderate volume and the growth of one or more dacite domes. Some radiocarbon dates on deposits of Pine Creek and Castle Creek age overlap, and if the two eruptive periods were separated by a dormant interval, it must have been short.
CASTLE CREEK ERUPTIVE PERIOD
The next period of activity marked a significant change in eruptive behavior and variety of rock types being erupted at Mount St. Helens. During the Castle Creek eruptive period, both andesite and basalt were erupted as well as dacite, and these rock types evidently alternated in quick succession. The overall sequence includes, from oldest to youngest, andesite, dacite, basalt, andesite, dacite, basalt.
Thus, the stratigraphic sequence of Castle Creek time is complex, and not all stratigraphic units are represented on all sides of the volcano. Northwest of Mount St. Helens, in the Castle Creek valley, the sequence preserved includes the following:
Lava flow of olivine basalt (youngest)
Lava flow of hypersthene-augite andesite
Tephra deposit of olivine-augite andesite scoria (layer Bo)
Pyroclastic-flow deposits of hypersthene-dacite pumice
Tephra deposit of hypersthene-augite andesite scoria (layer Bh)
Lava flow and lahars of hypersthene-augite andesite (oldest)
The pumiceous pyroclastic-flow deposits have a radiocarbon age of 2,000-2,200 yr. Deposits and rocks of Castle Creek age on the south and east flanks of the volcano include pahoehoe basalt lava flows whose radiocarbon age is about 1,900 yr, and pumiceous dacite tephra whose age is about 1,800 yr (layer Bi.). East of the volcano, layer Bi overlies a pyroclastic-flow deposit of pyroxene andesite, and directly underlies thin olivine basalt lava flows which probably are correlative with the uppermost unit in the Castle Creek valley. The Dogs Head dacite dome was extruded before those thin olivine basalt flows, probably during the Castle Creek eruptive period. Layer Bu is the youngest tephra of Castle Creek age; it underlies a deposit whose radiocarbon age is about 1,620 yr. This tephra is basaltic and probably was formed when thin olivine basalt lava flows were erupted near the end of the Castle Creek period.
Castle Creek time marked the start of eruptions that built the modern volcano. It is interesting to note that the change in eruptive behavior from that of the preceding 35,000-plus years did not follow a long period of dormancy like several that occurred during Mount St. Helens' earlier history. The dormant interval that followed Castle Creek time apparently lasted about 600 yr.
SUGAR BOWL ERUPTIVE PERIOD
During the next 1,200 yr, the only eruptions recorded at Mount St. Helens are those associated with the formation of Sugar Bowl, a dome of hypersthene-homblende dacite at the north base of the volcano. During extrusion of the dome, a directed blast carried rock fragments laterally northeastward in a sector at least 50 degrees wide and to a distance of at least 10 km. The resulting deposits are as much as 50 cm thick and consist of ash, lapilli, and breadcrusted blocks of dacite from the dome, fragments of charcoal, and stringers of material eroded from the underlying soil. A single fragment of charcoal from within the deposit has a radiocarbon age of about 1,150 yr, whereas a sample of wood charred and buried by the deposit has an age of about 1,400 yr (Hoblitt and others, 1980, p. 556). We provisionally assign an age of about 1,150 yr to the blast deposit; the older date may have been obtained from a fragment of a mature tree that was overridden by the blast.
A pyroclastic flow deposit of breadcrusted blocks, as well as prismatically jointed blocks of dacite of the same composition as the dome, was found on the north slope of Mount St. Helens downslope from Sugar Bowl; this pyroclastic flow may have occurred at the time of the lateral blast. Three lahars containing breadcrusted blocks of similar dacite were formerly exposed in the North Fork Toutle River valley west of Spirit Lake. These lahars may have been caused by melting of snow by the lateral blast or by the pyroclastic flow.
East Dome, a small dome of hypersthene-homblende dacite at the east base of the volcano, may have been formed at about the same time as the Sugar Bowl dome. East Dome is overlain by tephra of the Kalama period but not of the Castle Creek period, and could have been formed any time between the Castle Creek and Kalama eruptive periods, a time span of about 1,200 yr.
KALAMA ERUPTIVE PERIOD
Most of the rocks visible at the surface of the volcano before eruptions began in 1980 were formed during the Kalama eruptive period. Although the range in radiocarbon dates and ages of trees on deposits of Kalama age suggest that the eruptive period lasted from nearly 500 to 350 yr ago, all the events described here probably occurred during a shorter time span, perhaps less than a century.
The Kalama eruptive period began with the explosive eruption of a large volume of dacite pumice (layer Wn) which forms the basal part of tephra set W. Layer Wn was deposited northeastward from the volcano across northeastern Washington and into Canada (Smith and others, 1977, p. 209) and was followed by additional pumice layers. At about the same time, pyroclastic flows of pumiceous and lithic dacite moved down the southwest flank of the volcano. The relative timing of these events is poorly known because most of the air-falltephra was carried eastward and northeastward, whereas the pyroclastic flows have been found only on the southwest flank of Mount St. Helens.
A short time later, scoriaceous tephra of andesitic composition was erupted. In addition, andesite lava flows extended down the west, south, and east slopes of the volcano, and andesite pyroclastic flows moved down the north, west, and south flanks.
These eruptions of andesite were followed by the extrusion of the dacite dome that formed the summit of the volcano before the May 18, 1980, eruption. Avalanches of hot debris from the dome spilled down over the upper parts of the preceding lava flows, and some of this hot debris partly filled channels between levees of the andesite lava flows on the south side of the volcano (Hoblitt and others, 1980, p. 558). Late in this eruptive period, a pyroclastic flow of pumiceous dacite moved northwestward from the volcano down the Castle Creek valley and covered lahars of summit-dome debris. Charcoal from the pyroclastic-flow deposit has a radiocarbon age of about 350 yr (Hoblitt and others, 1980, p. 558).
The Kalama eruptive period was characterized by frequent volcanism of considerable variety; rock types being erupted alternated from dacite to andesite and back to dacite, and the volcano grew to its pre-1980 size and shape. The eruptive period was followed by a dormant interval of about 200 yr.
GOAT ROCKS ERUPTIVE PERIOD
The Goat Rocks eruptive period began about A.D. 1800 with the explosive eruption of the dacitic pumice of layer T. This pumice was carried northeast-ward across Washington to northern Idaho (Okazaki and others, 1972, p. 81) and apparently was the only eruptive product of that time. Many minor explosive eruptions of the Goat Rocks period were observed by explorers, traders, and settlers from the 1830's to the mid-1850's. The Floating Island Lava Flow (andesite) was erupted before 1838 (Lawrence, 1941, p. 59) and evidently was followed by extrusion of the Goat Rocks dacite dome on the north flank of the volcano (Hoblitt and others, 1980, p. 558).
The last eruption of the Goat Rocks eruptive period was in 1857, when "volumes of dense smoke and fire" were noted (Frank Balch, quoted in Majors, 1980, p. 36). A recent study of old records has suggested that minor eruptions of Mount St. Helens also occurred in 1898, 1903, and 1921 (Majors, 1989, p. 36-41). The published descriptions of these events suggest that they were small-scale steam explosions, and none produced deposits that were recognized in our studies.
DISCUSSION
One of the most interesting features of Mount St. Helens' history is the change in eruptive behavior that occurred about 2,500 yr ago. Eruptions of dacite had characterized the volcano for more than 35,000 yr. Then, with virtually no interruption in eruptive activity, andesite and basalt began to alternate with dacite, and not always in the same order. The chemical composition of eruptive products changed gradually during some episodes and abruptly during others. Thus, basalt followed dacite and dacite succeeded basalt; andesite followed dacite of considerably different SiO2 content, and vice versa. Some of these changes in composition of eruptive products are not adequately explained as results of eruption of cyclic sequences of compositionally different magmas derived from successively deeper levels in a larger magma body that differentiated at shallow depth, as proposed by Hopson (1971) and Hopson and Melson (19800. An alternative explanation that fits the stratigraphic record better, suggested by R.E. Wilcox (oral commun., 1974), is that some changes resulted from repeated contributions from more than one magma body, or from different parts of an inhomogeneous magma.
Explosive eruptions of volumes on the order of 0.1 to 3 km have occurred repeatedly at Mount St. Helens during some eruptive periods in the past. This record suggests that a similar sequence could occur during the present period of activity and could result in one or more explosive magmatic eruptions of similar or larger volume than the eruption of May 18. If the lengths of the last two eruptive periods are a valid guide to the future, we might expect intermittent eruptive activity to continue for several decades.
Eruptive History References
Alley, N.F., 1979, Middle Wisconsin stratigraphy and climatic reconstruction, southern Vancouver Island, British Columbia: Quatermary Research, v. 11, no. 2, p. 213-237.
Carithers, Ward. 1946. Pumice and pumicite occurrences of Washington: Washington Division of Mines and Geology Report of Investigations 15, 78 p.
Crandell, D.R., and Mullineaux, D.R., 1973, Pine Creek volcanic assemblage at Mount St. Helens, Washington: U.S. Geological Survey Bulletin 1383-A, 23 p.
_______ 1978, Potential hazards from future eruptions of Mount St. Helens volcano, Washington: U.S. Geological Survey Bulletin 1383-C, 26 p.
Crandell, D.R., Mullineaux, D.R., Miller, R.D., and Rubin, Meyer, 1962, Pyroclastic deposits of Recent age at Mount Rainier, Washington, in Short papers in geology, hydrology, and topography; U.S. Geological Survey Professional Paper 450-D, p. D64-D68.
Crandell, D.R., Mullineaux, D.R., and Rubin, Meyer, 1975, Mount St. Helens volcano; recent and future behavior: Science, v. 187, no. 4175, p. 438-441.
Fulton, R.J., and Armstrong, J.E., 1965, Day 11, in Schultz, C.B. and Smith, H.T. UY., eds, International Association (Union) of Quatemary Research Congress, 7th, 1965, Guidebook of Field Conference J., Pacific Northwest; p. 87-98.
Greeley, Ronald, and Hyde, J.H., 1972, Lava tubes of the /ave Basalt, Mount St. Helens, Washington; geological Society of American Bulletin, v. 83, no. 8, p. 2397-2418.
Hoblitt, R.P., Crandell, D.R., and Mullineaux, D.R., 1980, Mount St. Helens eruptive behavior during the past 1,500 years: Geology, v. 8, no. 11, p. 555-559.
Hopson, C. A., 1971, Eruptive sequence at Mount St. Helens, Washington: Geological Society of America Abstracts with Programs, v. 3, no. 2, p.138.
Hopson, C. A., and Melson, W. G., 1980, Mount St. Helens eruptive cycles since 100 A. D. [abs.]: EOS, v. 61, no. 46, p.1132-1133.
Hyde, J. H., 1975, Upper Pleistocene pyroclastic-flow deposits and lahars south of Mount St. Helens volcano, Washington: U.S. Geological Survey Bulletin 1383-B, 20 p.
Lawrence, D. B., 1939, continuing research on the flora of Mount St. Helens: Mazama, v.12, p. 49-54.
_______ 1941, The 'floating island" lava flow of Mount St. Helens: Mazama, v. 23, no. 12, p56-60.
_______ 1954, Diagrammatic history of the northeast slope of Mount St. Helens, Washington: Mazama, v. 36, no. 13, p. 41-44.
Mullineaux, D. R., 1974, Pumice and other pyroclastic deposits in Mount Rainier National Park, Washington: U.S. Geological Survey Bulletin 1326, 83 p.
Mullineaux, D. R., and Crandell, D. R., 1960, Late Recent age of Mount St. Helens volcano, Washington: U.S. Geological Survey Professional Paper 400-B. p. 307-308.
_______ 1962, Recent lahars from Mount St. Helens, Washington: Geological Society of America Bulletin, v 73, no. 7, p. 855-870.
Mullineaux, D. R., and Hyde, J. H., and Rubin, Meyer, 1975, Widespread late glacial and post glacial tephra deposits from Mount St. Helens, Washington: U.S. Geological Survey Journal of Research, v. 3, no. 3, p. 329-335.
Okasaki, Rose, Smith, H. W., Gilkeson, R. A., and Franklin, Jerry, 1972, Correlation of West Blacktail ash with pyroclastic layer T from the 1800 A. D. eruption of Mount St. Helens: Northwest Science, v. 46, no. 2, p. 77-89.
Smith, H. W., Okasaki, Rose, and Knowles, C. R., 1977, Electron microprobe analysis of glass shards from tephra assigned to set W, Mount St. Helens, Washington: Quaternary Research, v. 7, no. 2, p. 207-217.
Verhoogen, Jean, 1937, Mount St. Helens, a recent Cascade volcano: California University, Department of Geological Sciences Bulletin, v. 24, no. 9, p. 236-302.
Westgate, J. A., Smith, D. G. W., and Nichols, H., 1970, Late Quaternary pyroclastic layers in the Edmonton area, Alberta, in Symposium on pedology and Quaternary research, Edmonton, 1969, Proceedings: Alberta University Press, p. 179-187.