Chronology of the 1980 Eruptive Activity

Abstract

The climactic eruption began at 0832 PDT on May 18, probably triggered by an earthquake of magnitude 5 that caused failure of the bulging north flank as a 2.3-cubic-kilometer rockslide-avalanche. This failure rapidly unloaded the volcanic edifice, and probably caused the water in its hydrothermal system to flash to steam, initiating a series of northward-directed hydrothermal blasts that devastated an area of 600 square kilometers. These events in turn triggered a 9-hour dacitic magmatic eruption that drove a Plinian column more than 20 kilometers high, producing ash fallout for more than 1,500 kilometers to the east as well as pumiceous ash flows on the volcano's north flank. Catastrophic mudflows and floods were generated from rapid melting of snow and ice and water derived from the avalanche.

Smaller but significant magmatic eruptions occurred on May 25, June 12, July 22, and October 16-18; each lasted as long as several hours and produced eruption columns more than 10 kilometers high, dacitic fallout, and pumiceous ash flows. A dacitic dome emplaced during or after the June eruption was partially reamed out in the pyroclastic eruption of July 22. Another dome emplaced after the August eruption was removed and then replaced by a third dome during the mid-October eruption. During the times that they remained intact, each of these domes appeared to act as a leaky plug to gas emissions through the volcano's central crater vent. Occasional violent gas emissions between major eruptions produced brief eruption plumes to heights of a few kilometers. Beginning in late December and continuing through the end of 1980, a new dome was emplaced and grew rapidly to the accompaniment of increased seismicity and gas emissions.

Introduction

Acknowledgments

Among the many contributors of data, none was more essential to the systematic reconstruction of the events of 1980 at Mount St. Helens than David Johnston, to whose memory this report is dedicated. Dave, who was present through all of the activity up to the climactic eruption and who lost his life in that eruption, provided far more than data. His insights and his thoroughly scientific attitude were crucial to the entire effort; they still serve as a model for us all.

Initial Period of Seismic and Steam-Blast Activity

The first major event of the 1980 activity was an earthquake of magnitude 4.0+ at 1547 PSTl on March 20 (day 80:23:47 UT). The swarm of earthquakes increased rapidly to a climax in the late afternoon of March 25, when 24 earthquakes of magnitude 4 and greater occurred during an B-hr period. Overflights on March 24 had indicated no major changes in the appearance of Mount St. Helens other than avalanches of snow and ice. Additional overflights on March 25, during the peak of seismic activity, revealed several new fractures through glaciers high on the mountain and numerous additional large rockfalls and ava- lanches. None of the new fractures, however, coincided with the larger fractures that later formed across the summit area; the fractures that formed on March 25 probably resulted directly from ground shaking and the accelerated downslope movement of glacial ice on the volcanic edifice. After March 25, seismicity declined somewhat but remained at a high level, with about 30 events per day having a magni- tude of 3 or greater (6/day of magnitude 4 or greater).

The first eruption occurred in the early afternoon of March 27. Although extensive cloud cover had hidden the volcano from the air since the morning of March 25, a loud boom was widely heard at 1236 PST on March 27. Aerial observers reported a dark dense column of volcanic ash rising through the clouds, eventually to a height of 2,000 m above the volcano.

With clearing weather later in the afternoon, several changes were conspicuous on the mountain. A new crater about 60-75 m across had formed in the northern part of the old 400-m-wide ice-filled summit crater, and snow on the southeast sector of the volcano was covered by dark ash emitted from the new crater (fig. 6). The summit area was bisected by an east-trending fracture nearly 1,500 m long that extended from high on the northwest flank, across the old crater, down the upper northeast flank. Another less continuous fracture system paralleled this master fracture just north of the old crater rim and bounded the south side of a newly uplifted block, or bulge, on the volcano's north flank. These changes clearly had occurred during the period of extremely high seismicity and initial eruption, between observations on the morning of March 25 and the afternoon of March 27. One observer, David Gibney-an aerial spotter of the U.S. Forest Service (oral commun., 1980)-reported seeing the large fractures open and close and the uplifted north flank continue to break and rise during the few hours after the first eruption.

A second explosive eruption, beginning at about 0200 on March 28 and lasting nearly 2 hr, was observed from the air. Ash from that eruption spread for many kilometers to the east of the volcano. By nightfall on March 28, at least a dozen more eruptions had occurred, many of them lasting only a few minutes but some for nearly an hour. Poor weather hampered observations on all of these days. By March 29, however, clear views of the summit revealed the presence of a second larger crater west of the one first seen on March 27 (fig. 21). A septum about 10 m wide separated the two craters. Pale-blue flames were first observed on the night of March 29 and were subsequently observed in each of the two craters.

On April 1, a weak burst of harmonic tremor lasted for about 5 min, but stronger tremor bursts the next day were recorded by seismometers as far away as 100 km. Sporadic harmonic tremor continued until April 12, but these tremor bursts could not be correlated directly with the visible character or intensity of eruptive events. The frequency, duration, and intensity of eruptive blasts gradually decreased until April 22, when eruptions temporarily ceased. Eruptions had declined in frequency from an average of about Ilhr in March to about II day by the end of this first period of activity.

The eruptions of this first period produced only lithic-crystal ejecta, composed of fragments of preexisting rocks. All of this material was emitted from vents in the new summit craters, which were repeatedly reamed. By April 7, the septum separating the two craters had broken down entirely, and the enlarged single crater grew to about 500 m from west to east and about 300 m north to south (fig. 7). By late in the month, the crater was about 100-250 m deep. (Its rim was highly irregular in elevation.)

As eruptive activity declined slowly through early and mid-April, earthquakes continued at still impressive rates, generally more than 301 day of magnitude 3 or greater. Many of these earthquakes were large enough to be felt strongly on and immediately adjacent to the volcano. All were very shallow, however, and few were felt very far from the volcano. The epicenters were confined to a small area that coincided with the uplift or bulge on the volcano's north flank. These frequent shallow earthquakes triggered numerous avalanches, which were concentrated around the northern sector of the mountain. Most avalanches started on the upper slopes of the volcano where ashfall was heaviest; because they involved dark ash-laden snow, the avalanche deposits stood out prominently and were at first mistaken for mudflows.

After eruptive activity ceased temporarily on April 22, fumaroles vented continuously within the crater, in contrast to earlier conditions, when the crater generally was clear between eruptions. The most continuous and generally largest fumarole was on the south wall just above the eastern part of the crater floor. The fumaroles could not be sampled because of frequent ice avalanches from the crater wall above, but a water sample was collected from a small pond on the eastern crater floor. Chemical analysis of the water indicated high chloride, sulfate, and ammonia, relative to local snowmelt, suggesting a contribution to its composition by fluids of deeper origin.

Small eruptions resumed on May 7 and continued sporadically through May 14; these eruptions were similar to those of March and April. A general description can be applied to all of the pre-May 18 eruptions. In a few events, columns of ash and condensed steam rose as high as 3,000 m above the crater floor. Generally, however, the gas-driven bursts of ash-laden material rose less than 500 m above the vents, above which condensed steam billowed in white clouds. Collected samples of distal ejecta consisted entirely of lithic-crystal ash; at the crater rim, the ejecta also included abundant blocks of dacite from the summit dome and sparse blocks of more mafic rock types from the volcanic edifice. The eruptions consisted of intermittent blasts at intervals of a few minutes to several hours that varied greatly in size and duration. The smaller eruptions were confined to the crater; some consisted of only a single blast. Larger ones lasted as long as several hours, with individual pulsating blasts occurring up to several times per minute, emitting clouds that reached 1,500-3,000 m above the crater. Varying winds carried ash to all flanks of the volcano, entirely mantling its upper slopes.

Close examination of especially the smaller eruptions showed many of them to consist of three parts: (1) a lower, fingerlike ash column, (2) a gray-brown, ash-laden cloud above, and (3) an upper white cloud. The fingerlike ash column was nearly black and had spiked tops similar to the II cockstail" eruptive columns commonly described elsewhere from steamblast eruptions (fig. 8). These fingerlike columns rose vigorously from the crater bottom; several clearly rose through small, temporary crater lakes and others through collapse pits in the ice-block talus on the crater walls and floor. Many of these columns were dark probably because the ash they carried was wet; ash on the surface became lighter after it dried out over several days. Other evidence suggests that the columns had a high water content: many of them were emitted from lakes; large amounts of water fell from the columns and coursed down the crater slopes; and new lakes commonly formed or preexisting lakes were fullest immediately after such blasts.

Roiling, gray-brown, ash-laden clouds developed from the tops of the ash-finger columns and commonly obscured them (fig. 9). These clouds expanded rapidly as blocks of rocks and ice trailing white vapor-condensate were blasted through them. The upper white clouds were formed by continued expansion of the gray-brown ash cloud and by the condensation of water vapor, growing on calm days to great heights. Under normal windy conditions, however, they blew nearly horizontally, dropping thin, nearly vertical veils of ash.

Downslope winds commonly generated sweeping clouds of ash down the volcano's flank. Although these ground-hugging clouds were first regarded as possible pyroclastic flows, further observations showed them to be driven by the wind rather than by the eruption column itself. Further movement down- slope, however, apparently occurred as density flows. Typically, the ash swept up in rising air currents from the lower flanks of the volcano. The deposits left by these flows commonly were very thin, some of negligible thickness; where conspicuous, their downslope limits were marked on new snow by a sharp scalloped boundary of black ash. Occasionally these dust-laden density currents in their downward sweep produced spectacular displays of lightning parallel to the volcano slope.

Fumarolic activity continued after eruptions resumed in early May, and it appeared to increase gradually. A large pitlike steam vent opened high on the Shoestring Glacier about May 11 and remained open and active during the succeeding week. More new steam vents appeared to open in the upper part of the bulging north flank during the first half of May, and thermal infrared scanning at about mid-month revealed numerous areas of thermal emission in the ice crevasses of that area.

The prominent topographic bulge that had been noted on the north flank of the volcano on the first day of eruption continued for nearly 2 mo to become larger and more conspicuous. Ground ruptures at the 5,400-ft level on the north flank were first noted on April 3 and provided early evidence that the bulge affected a large part of the cone. The first detailed photogrammetric measurements, completed in midApril and compared with contours based on photography of August 1979, revealed the startling dimensions of this bulge. By May 12 (the date of the last contour map predating the climactic eruption of May 18), the high point north of the old crater rim stood 150 m above a corresponding point on the former north slope, and the Goat Rocks area low on the bulge had been displaced northward by 106 m. Geodetic measurements showed that displacement was only slightly upward near the top of the bulge; it was mainly outward-nearly horizontally-at consistent rates of about 2 m/day. No appreciable change oc- curred in this rate of displacement, even up to 1 1 /2 hr before the climactic eruption. Photogrammetric evidence suggests that no appreciable bulging of the north flank occurred before the events of 1980. Comparison of maps made from aerial photographs of 1952 and 1979 indicates that if any bulging did occur before August 1979, the amount was close to the resolution possible with the BO-ft (24.4-m) contour interval of the maps-an order of magnitude less than that which occurred between late March and mid-May 1980.

During late April and early May, the upper part of the bulge changed in general appearance. A high point that had formed in late March (north peak 1) just north of the crater stagnated and subsided during April as a second high point (north peak 2) began to form farther north (fig, 10). An east-west line between these two points appeared to be a fracture zone that delimited areas of differing rates and styles of deformation. In early May, old fractures and disturbed segments of the north and west rims of the summit crater south of the fracture zone were partly filled by snow and drifted volcanic ash and appeared to be parts of a coherent graben block that moved downward with the crater. By contrast, the surface of the actively bulging main part of the north flank continued to break and distort, indicating persistent internal deformation as well as outward bodily displacement.

By mid-May, earthquakes of magnitude 3 or greater continued to occur at a rate of 20-40/ day, including 5-10 of magnitude 4 or greater. Two magnitude-5,O earthquakes occurred, on May 8 and 12. On May 8, the day following a resumption of eruptions, two periods of harmonic tremor were recorded, each lasting only a few minutes. No further harmonic tremor occurred until after the beginning of the May 18 eruption.

In summary, throughout an initial 2-mo period, ejecta consisted entirely of fragmental material derived from the volcanic edifice, most of it generated by shattering and pulverization at shallow levels within the 350-yr-old summit dome. A moderate amount of this ash was distributed 50 km away and some was reported as far as 100 km to the east, but most fell within a few kilometers of the volcano's summit. Evidence for the involvement of water was abundant, and the eruptions apparently all resulted from a steam-blast mechanism that reflected internal heating of the volcano by a shallow intrusion that also produced localized, but very high, seismicity and a rapidly and continuously bulging north flank.

The Climactic Eruption

At 0832 PDT (day 139:15:32 UT), with no known immediate precursors, a magnitude-5 + earthquake triggered a rapid series of events. As seen by Keith and Dorothy Stoffel (written commun., 1980) from a small aircraft at low level directly above the summit crater, the earthquake caused avalanching from the walls of the crater and, only a few seconds later, triggered a sudden instability of the north flank. The entire north flank was described as II quivering ll and appeared to almost liquefy. The slope failed along a surface intersecting the northern of the two high points on the north flank, near the east-west fracture separating the active bulge from the crater block. As the north flank began to slide away from this surface, a small, dark, ash-rich eruption plume rose directly from the base of the scarp and another from the sum- mit crater rose to heights of about 200 m. As virtually the entire upper north flank slid off the cone and became a massive debris avalanche, a blast broke through the remainder of the flank, spewed ash and debris over a sector north of the volcano (fig. 11), overtook the massive avalanche, and devastated an area nearly 30 km from west to east and more than 20 km northward from the former summit of the volcano. In an inner zone extending nearly 10 km from the summit, much of which had been densely forested, virtually no trees remained. Beyond, nearly to the limit of the blast, all standing trees were blown to the ground, and at the blast's outer limit the trees were left standing but thoroughly seared. The devastated area of 600 km 2 was blanketed by a deposit of hot debris carried by the blast.

The sole of the debris avalanche was nearly at the base of the steep volcanic cone on the north side; the avalanche moved down the lower gradients of the volcano's outer flank and was nearly blocked by a ridge 8 km to the north. Part of the avalanche rounded the east end of that ridge and displaced the water from Spirit Lake, raising the bed in its southern part by more than 60 m. The bulk of the avalanche, however, turned westward down the valley of the North Fork Toutle River to form a craggy and hummocky deposit, part of which crossed the ridge to the north, but most of which flowed as far as 23 km down the North Toutle. The total volume of the avalanche in place is about 2.8 km 3 , and its length makes it one of the largest on record.

Water incorporated by the avalanche from the North Fork Toutle River and possibly from Spirit Lake combined with melting blocks of ice from the torn-out glaciers of the volcano's north flank and melting snow and ice from the volcano's remaining slopes to produce mudflows that later in the day coursed across the avalanche and down the North Fork Toutle River, sweeping up thousands of logs from timbering operations in the valley and destroying most bridges across the river. The mudflows continued downstream, depositing sediment in the Cowlitz River channel and also obstructing the deep-water navigation channel of the Columbia River. Smaller mudflows were produced from the east flank of the volcano and went down the valleys of Muddy River and Pine Creek into Swift Reservoir (pI. 1). Yet other mudflows and floods went from the volcano's northwest flank down the South Fork Toutle River; smaller floods occurred in the Kalama River on the southwest.

The initial events of the eruption-the rockslide avalanche, the northward blast of ash and debris, and the mudflows-caused most of the casualties and destruction in the immediate region of the volcano. However, within a few minutes a Plinian eruption column (fig. 12) began to rise from the position of the fonner summit crater and within less than 10 min had risen to a height of more than 20 km. Ash from this eruption cloud was rapidly blown east-northeastward, producing lightning and starting hundreds of small forest fires, causing darkness eastward for more than 200 km, and depositing ash for many hundreds of kilometers. Major ash falls occurred as far east as centra] Montana and ash fell visibly as far eastward as the Great Plains of the Central United States, more than 1,500 km away. As this Plinian eruption column grew, it reamed out the volcanic conduit. The eruptive crater, along with the upper 300 m of the cone that was entirely removed by the initial slide and blast, fonned a great amphitheater 1 1 /2 x 3 km across, enclosed by the volcano's former east, south, and west flanks (fig. 16).

The Plinian phase of the eruption continued vigorously for 9 hr and produced numerous ash flows. Some of these were thin flows that spread out over much of the upper surface of the volcano and were generated by fallback from the expanding eruption column in the vicinity of the summit. Most of the flows were directed out through the large northward breach of the crater to form a fan of pumiceous ash flows over the avalanche, extending to Spirit Lake and part way down the North Fork Toutle River valley. Ash flows continued to be emplaced at least until dark on May 18. The hot blast deposits, the avalanche, and these ash flows were frequently disrupted in the vicinity of Spirit Lake and its former drainage into the North Fork Toutle River by large secondary steam-blast eruptions that formed craters as large as 20 m across and drove columns of ash to heights as great as 2,000 m above the surface,

The Plinian eruption began to decrease in intensity at about 1730 on May 18 and by the next morning had decreased to a very low level. Minor eruptive emissions persisted almost continuously until May 21 from a smaller vent crater near the center of the large amphitheater, but observations of their size and character were restricted by poor weather. Weak steam and ash eruptions continued intermittently for about a week.

The magma tapped during the major eruption of May 18 was dacitic. The principal juvenile material was a light-colored hypersthene-hornblende dacitic pumice having a range of silica content of about 63-64.5 percent. A texturally different rock type-the first magmatic material erupted-was incorporated into the initial blast deposits. It was a denser, and therefore, darker dacite whose composition is at the silicic end of the range of the pumice compositions. A total of about 0.2 km 3 of magmatic material (as reduced to the equivalent volume of dense rock) was erupted on May 18,

In summary, although no precursory changes in seismicity or in the rate of deformation on the north flank had provided immediate warning of an impending major slope failure, the longer term likelihood of such failure suggested the possibility that it could trigger a major volcanic eruption. The bulge probably had nearly reached the point of instability and might have been almost ready to begin creeping more rapidly toward failure when a magnitude-5 + earthquake at 0832 on May 18 intervened to push the mass over its stability threshold. The earthquake triggered the rockslide-avalanche, which in turn unloaded the hydrothermal system in the volcano that had driven intermittent steam-blast eruptions for the previous 2 mo. This abrupt release of confining pressure on hot water in the system caused a massive flashing to steam, initiating a hydrothermal blast that was directed laterally through the landslide scar. This sequence of events in turn caused further unloading of magma in the shallow body that was intruding the volcano, causing it to de-gas and to drive the Plinian eruption column that continued for the next 9 hr.

Period of Subsequent Pyroclastic Activity and Lava Domes

For the next 2 1 /2 weeks, the volcano continued to emit large quantities of gas that rose in plumes of steam condensate to altitudes of 3-5 km above sea level. Sulfur emissions, monitored since late March, had remained low until the eruption of May 18. After that eruption, sulfur gases were emitted at a higher rate, but relatively little ash was carried in the gas plumes and appreciably none fell more than a few kilometers beyond the volcano. During this time, no lava appeared at the surface, but there were several night observations of incandescent rock, probably caused by hot gases streaming through the vents from a magma body not far below, Also during this time, small steam blasts continued to erupt through the avalanche and ash-flow fill in the former North Fork Toutle River valley north of the volcano. Some of the craters formed by those eruptions were enlarged con- siderably by repeated blasts and by the coalescence of smaller craters.

A third magmatic eruption took place on June 12. This eruption was preceded by several hours of harmonic tremor that began around midday and gradually grew in intensity throughout the afternoon. A marked increase in tremor amplitude was noted at 1905 PDT, and an eruption drove an ash column to at least 4 km above sea level by 1910. Tremor amplitude decreased markedly immediately after this brief eruptive pulse and fluctuated at moderate levels for more than 2 hr. The temporary lull was broken by a rapid, large increase in tremor amplitude at 2111, and by 2118 an eruption column had risen to 15 km above sea level. The height of the column fluctuated between 5 and 12 km above sea level until 0043 on June 13, when it decreased abruptly.

Prevailing winds carried the ejected ash south-southwest, allowing centimeter-sized pumice fragments to fall in Cougar, Wash., about 16 km downwind. Portland, Oreg., and Vancouver, Wash., received moderate ash falls beginning at about 2250.

Several new ash flows issued from the vent crater and descended the volcano's north flank during the June 12 activity. These flows nearly reached the south shore of Spirit Lake and overrode and locally ponded within steam-blast craters in the ash flows of the May 18 eruption. Like the May 25 deposits, the June 12 material contained some dark-gray, dense pumice as well as lighter colored and more vesicular pumice.

A lava dome (fig. 13) probably began to rise in the vent crater shortly after the explosive eruptions of June 12, but because of poor visibility it was not sighted until June 15. Several features suggest that the dome was close to the surface before and during the June 12 eruption. At the time of the last previous good view of the crater amphitheater on June 11, a large elliptical lake about 300-400 m long and 50-75 m wide occupied much of the southern crater floor. Its elliptical shape resulted from a broad flat topographic high in the center of the crater at the location of the subsequent dome. This mound, com- pletely mantled by gray ash, may have represented the early rise of a magma plug nearly to the crater floor. Also, in contrast to the ash flows erupted on May 18 and 25, the ash flows of June 12 contain not only pumice but also abundant blocks of dense, gray dacite, as much as 2 m in diameter. The June 12 deposits thus resemble "block-and-ash" flows emplaced as a result of dome collapse on other volcanoes and may contain parts of the solidified margins of a crypto dome prior to its arrival at the surface.

Repeated observations from a hovering helicopter documented steady dome growth at a rate of 2-3 m/day from June 15 through the late afternoon of June 19. No observations were possible during poor weather from June 20 through June 27, but observations on June 28-29 suggested a stabilization after June 19 or even a slight collapse.

In mid-July, measurements showed an apparent northward expansion of the rampart north of the vent crater of about 5 cm/day. Mount St. Helens again erupted at 1714 PDT on July 22 (fig. 14) after a small swarm of shallow local earthquakes that began about 1000 that morning. From 1400 to 1500 there were 4 earthquakes, followed by 9 more from 1500 to 1600, and by 20 from 1600 to 1700. The first ash eruption lasted about 6 min and a pyroclastic column reached 14 km above sea level. A second ash cloud erupted at 1825 and lasted about 22 min, reaching a height of more than 18 km. The third and longest ash eruption began at 1901 and lasted about 2 hr and 40 min (fig. 14); maximum column heights fluctuated (the highest at 14 km at 1907) and from a distance gave the appearance of several separate eruptions. Winds blew the main ash cloud northeastward, and satellite images showed it to cross from Idaho into Canada. Pumiceous ash flows were erupted on the north flank of the volcano, especially during the second and third eruptive bursts. No harmonic tremor accompanied the earthquake swarm prior to the 1714 eruption, but tremor did occur during the periods of ash emission.

The July 22 eruption blasted a new crater through the center of the June lava dome. The walls of the new crater, which had a diameter of about 300 m and a tail-like southward extension of about 400 m, were in the still-hot June dome, which was mantled by several meters of new ash from the July eruption. No new dome was produced, and this new crater remained intact until an eruption in early August. Following the explosive eruption of July 22, the crater area fumed-copiously at times-and columns of condensing water vapor rose to elevations as high as 3,000-3,500 m above sea level. A dull red in- candescence was observable in dim light on the walls of the vent crater. The walls of the larger amphitheater around the vent crater produced numerous rock falls and avalanches. Once again, measurements in early August suggested northward expansion of the northern rampart of the vent crater by one to several centimeters per day. The principal measured gas emissions, CO 2 and S02, were moderately high at the end of July but decreased somewhat in early August, and the ratio CO 2 : S02 declined markedly on August 6.

Just after noon on August 7, harmonic tremor began, and it continued to increase in amplitude for several hours. The combination of changing gas fluxes, harmonic tremor, and occasional small earthquakes suggested an impending eruption, and ash emission started at 1623, growing rapidly to full eruption by 1627. This first burst produced an ash-laden eruptive column that rose to an elevation of more than 13 km. A small pumiceous ash flow swept the area below the breach on the north side of the amphitheater (fig. 15). The flow reached part way to Spirit Lake, leaving a thin lobate deposit. Smaller eruptions continued through the late afternoon and evening, with one major sequence around 1930. A culminating burst, nearly as large as the first one, began at 2232. The intensity of harmonic tremor decreased and small deep earthquakes occurred during the period following this last eruptive burst, which ended before midnight.

A dome began to rise in the vent crater on the morning of August 8, filling it to about half its former depth of 90 m by the end of the day and stopping growth just below crater-rim level by August 10. For several weeks following, the volcano remained relatively quiet; there were few earthquakes and no major eruptions. Some incandescence could still be seen in the walls of the vent crater and in cracks on the surface of the new lava dome. Gas emissions fluctuated from moderate to low levels.

Following the August 7 eruption, the volcano generally remained very quiet for more than 2 mo (fig. 16), and seismicity dropped to its lowest levels since before the reawakening of the volcano in March. A slow outward movement of the unstable sector on the north flank, however, indicated that magma within the volcano remained active. During the second week of October, small shallow earthquakes resumed at the rate of one to a few per day. On October 16, their frequency gradually increased to several per hour, and at 1902 PDT a magnitude-3 earthquake occurred. The small earthquakes continued, and a pyroclastic eruption began at 2158, lasting between 5 and 10 min. Shallow seismicity continued, and other eruptive pulses followed at 0928 and 2112 on October 17 and 1235 and 1428 on October 18. Each of these events lasted from about 5 min to nearly an hour, although the most vigorous activity during each of them took place in the first few minutes. The pyroclastic eruption columns reached maximum heights of about 14 km. Small pyroclastic flows occurred during two of the events, descended the north flank of the cone, and reached lengths of about 5 km. The wind was variable throughout these events but blew generally from a sector between northwest and northeast; light ash falls occurred in a southern sector extending from the volcano to the Portland- Vancouver area and as far east as The Dalles, Oreg. Following the final pyroclastic eruption on October 18, a new lava dome emerged from the floor of the crater and grew at a dramatic rate through the afternoon and evening. By the following morning it had obtained its ultimate size of more than 200 m in diameter and 40 m in height.

During the intervals between the major eruptions after May 18, occasional emissions of gas and some ash broke the general quiet; the ash probably was entrained, being derived from ejecta in and near the vent crater. Some of these events were accompanied by brief pulses of harmonic tremor and produced small perforations in the domes or crater floors. Significant events of this type occurred on July 28, August 15, and several times in September and October. After the mid-October eruption, the volcano returned to a very quiet state that persisted until late November, with very few earthquakes, rare episodes of harmonic tremor, little ground deformation, and only slow rates of gas emission. Beginning in late November, a new pattern began to develop. Low-level seismic events of uncertain character preceded or accompanied episodes of low-amplitude harmonic tremor. Accompanying some of these events were weak to moderate gas emissions, some of which carried entrained ash. On December 13, a small crater formed and a wedge-shaped sector of the dome was blasted away in one of these gas-emission events. Beginning during a period of increased seismicity between December 25 and 28, a new dome was extruded alongside the October dome. The new dome continued to rise and expand rapidly through the end of 1980, accompanied by frequent small, shallow earthquakes, harmonic tremor, and heavy gas emissions.

During the period of intermittent volcanic and seismic activity that followed the climactic eruption of May 18, there were significant adjustments of the drainage system to the changes wrought by that eruption. The drainage that was disrupted and blocked by the massive avalanche of May 18 quickly began to reestablish itself across the hummocky surface of the avalanche in the North Fork Toutle River valley. The river course along the lower part of the avalanche had already been reestablished on the south side of the avalanche before the end of the day on May 18. One event illustrates in part how the establishment and integration of drainage continued to occur through the summer months. During the last 2 weeks of August, water from Maratta Creek, which had been impounded by an avalanche levee on the north side of the valley, was released suddenly onto the surface of the avalanche and filled a small depression. The water was retained temporarily in a deep pond a few kilometers above Camp Baker. On the afternoon of August 27, the rising pond overtopped its rim and eroded its outlet as it drained downvalley. Much of the water was retained by a check dam under construction by the U.S. Corps of Engineers, but some spilled over, destroying some temporary bridges and portions of access roads along the North Fork Toutle River as far downvalley as the town of Toutle.

In summary, a possible pattern of volcanic behavior may be evident in the post-May 18 events at Mount St. Helens. A magma column seems to be slowly and intermittently rising; sometimes it is entirely confined beneath the surface and at other times its upper portion emerges as a lava dome. The outer solidified shell of this magma column is an imperfect seal, and fluctuating amounts of gas continuously escape to the atmosphere. At irregular intervals, pressure within the column exceeds the confining strength, and gas is released violently, blasting out new craters and giving rise to pyroclastic flows and to high eruption columns that deposit new ash. Renewed rise of the magma column after each eruption replugs the vent and may produce a new dome.