Extrusion of a new lobe onto the NE portion of the composite lava dome started 6 September and had ended by noon on 11 September, after reaching a volume comparable to previous lobes. The eruption was preceded by increases in the crater displacement rate, SO₂ emission, and seismicity. This was the first extrusion episode at Mt. St. Helens in which weather conditions allowed observation of the crater immediately before and during the event.

On 5 July, the USGS installed three continuously recording bubble tiltmeters along a roughly N-S line within 150 m of the E side of the composite dome. The central tiltmeter, directly E of the middle of the dome, began to record gradual inflation on 8 July. The rate of inflation recorded by this tiltmeter increased systematically through July and most of August, reaching 235 µrad/day by 27 August and totaling about 2,800 µrad for that month. Telemetry problems plagued the tiltmeter 175 m to the NE (at the same site as the instrument destroyed by a rockfall during the June extrusion). Field measurements of its output in July and data returned after resumption of telemetry 13 August showed a similar trend of uplift, but at only about half the rate recorded by the central tiltmeter. Before another rockfall interrupted telemetry 29 August, the N tiltmeter had detected about 800 µrad of inflation in 16 days. About 300 m SW of the central tiltmeter (directly E of the S end of the dome), the third tiltmeter, although functioning properly, recorded no significant net tilt through August.

Of the numerous thrust faults in the crater floor around the dome, the most vigorous showed rapid acceleration in late August. The highest measured rate of movement occurred about 100 m SW of the dome, along a fault first observed after the June extrusion. From 1.8 cm/day between 18 and 22 August, thrusting along this fault increased to 6.3 cm/day 27-29 August. Two other thrusts W of the dome showed similar accelerations but less total movement. Near the very active central tiltmeter E of the dome, movement resumed along an older thrust 22 August and accelerated on the 29th.

However, most of the thrust faults that formed prior to the June extrusion remained inactive. The substantial differences in rates of thrust fault movement, combined with the variation in trends shown by the three bubble tiltmeters, indicated to USGS personnel that the crater floor was behaving as a group of independent blocks or plates on a scale of the order of 100 m rather than as a single coherent body.

Rates of displacement along the former N-crater rampart varied considerably during August. At the most active site, the rate of outward movement increased to 2 cm/day between 12 and 18 August, declined to about 0.7 cm/day until late August, then resumed an irregular acceleration. A very gradual increase in the number of shallow volcanic earthquakes began about 29 July. Most of the events were centered near the former site of Goat Rocks Dome, slightly NW of the present composite dome. Several larger events occurred 13-16 August, then seismic activity remained relatively constant until early September. Small steam-and-ash plumes were occasionally ejected during August, sometimes accompanied by brief periods of harmonic tremor. The volume of SO₂ emissions, measured by COSPEC from fixed-wing aircraft flying under the plume, averaged 60 t/d 9-18 August, then increased sharply to about 360 t/d 19-24 August.

Increases in rates of tilting, outward movement of the former N crater rampart, movement along thrust faults, and SO₂ emission, prompted a USGS advisory on 26 August that stated in part "an eruption, probably of the dome-building type, will likely begin in 1-3 weeks." After this advisory was issued, activity in the crater continued to increase, although rates of SO₂ emission dropped to an average of 150 t/d 26 August-5 September. Between 3 and 5 September, outward movement of the former N crater rampart reached 10.4 cm/day at one site and 6.5-7 cm/day at three others. Thrusting along the most active fault about 100 m SW of the composite dome accelerated to 23 cm/day 1-3 September, 28 cm/day 3-4 September, and 48 cm/day 4-5 September. Movement along the older thrust fault near the central tiltmeter reached a rate of 7-8 cm/day by 4 September.

Beginning 2 September, USGS personnel working in the crater noted 1-2 rockfalls per hour and frequent audible and felt earthquakes. However, the earthquakes were probably very shallow, as no significant increase in seismicity was recorded by the University of Washington seismic net through 4 September. Audible and felt earthquakes in the crater were nearly constant on 5 September, and rockfalls increased further, particularly from the overhanging NE portion of the June lobe. Recorded seismicity began to increase shortly after noon, and increased more rapidly during the predawn hours of 6 September, triggering a joint USGS-University of Washington advisory at 0800 on 6 September that predicted a dome-building eruption within the next 12-48 hours.

During this period, sharply varying data were returned by the three continuously-recording tiltmeters. After recording about 80 µrad/day of inflation 1-4 September, tilt at the N station reversed to relatively slow deflation on 5 September. Deflation continued on this instrument until its telemetry was ended by a rockfall during the afternoon of 6 September. Only 175 m to the SE, the central tiltmeter continued to record increasingly vigorous inflation, with rates reaching 700 µrad/hr on 6 September. This instrument recorded more than 10,000 µrad of inflation on 6 September before an incandescent boulder ended its telemetry during the afternoon. The S tiltmeter (about 300 m SW of the central instrument) had recorded no significant tilt previously but began to show deflation 5 September that continued through the 8th.

The seismic character changed to lower frequency events with emergent arrivals after dawn on 6 September, and epicenters moved to the area of the present dome. At about 1000 on the 6th, avalanche events began to dominate the seismic record, with only a few discrete low-frequency events appearing for the next several hours. USGS personnel working in the crater observed huge blocks falling from the NE portion of the June lobe, and were soon forced to retreat to a ridge N of the crater. Avalanche events peaked on the seismic record at about noon, but remained at high levels until about 1700. Clouds of dust from the frequent rockfalls made observation of the crater difficult, but by 1500-1530 it was evident to USGS personnel that the entire NE portion of the June lobe was breaking up. A bulge appeared to be developing on the E side of the lobe, but poor viewing conditions made this observation uncertain. By 1600-1700, an area of tens of cubic meters of fresh lava was clearly visible on the dome, and by 1830 many glowing rockfalls could be seen. . . . The number of seismic events began to decline after 1700. Significant numbers of low-frequency events resumed briefly about 2200, but seismicity dropped sharply at about 2330.

Aircraft crews monitoring the crater during the night of 6-7 September saw numerous glowing rockfalls, and by 0500 on 7 September a new lobe had been emplaced in the area formerly occupied by the NE portion of the June lobe. Most of the NE portion of the June lobe had fallen as talus, but from its high point to its SW margin, the June lobe remained intact. Slow aseismic growth and downslope spread of the new lobe continued through the afternoon of 10 September, but USGS field parties reported that growth had stagnated by noon 11 September. The new lobe was comparable in size to lobes extruded in previous episodes, but precise determination of its volume and daily growth rate await analysis of airphotos and reduction of field data.

Geological Summary. Prior to 1980, Mount St. Helens was a conical volcano sometimes known as the Fujisan of America. During the 1980 eruption the upper 400 m of the summit was removed by slope failure, leaving a 2 x 3.5 km breached crater now partially filled by a lava dome. There have been nine major eruptive periods beginning about 40-50,000 years ago, and it has been the most active volcano in the Cascade Range during the Holocene. Prior to 2,200 years ago, tephra, lava domes, and pyroclastic flows were erupted, forming the older edifice, but few lava flows extended beyond the base of the volcano. The modern edifice consists of basaltic as well as andesitic and dacitic products from summit and flank vents. Eruptions in the 19th century originated from the Goat Rocks area on the N flank, and were witnessed by early settlers.

Information Contacts: T. Casadevall, D. Dzurisin, C. Heliker, USGS, Vancouver, WA; C. Boyko, S. Malone, E. Endo, C. Weaver, University of Washington; R. Tilling, USGS, Reston, VA.