FIRE and MUD: Eruptions and Lahars of Mount Pinatubo, Philippines

Ground Deformation Prior to the 1991 Eruptions of Mount Pinatubo

By J.W. Ewert,1 Andrew B. Lockhart,1 Sergio Marcial,2 and Gemme Ambubuyog 2

1U.S. Geological Survey.

2Philippine Institute of Volcanology and Seismology.


Measurements of ground deformation during several weeks prior to the climactic eruption of Mount Pinatubo came from two telemetered tiltmeters located on and near the volcano and from a small quadrilateral array of points across the chain of craters that opened on April 2, 1991. Overall, the deformation data are sparse, owing to logistical constraints and hazards considerations. Data from one of the tiltmeters provided information on the timing of edifice inflation and on the development of a magma conduit to the surface. On June 7, 1991, tilt and seismic data were used to reach the conclusion that magma had reached very shallow levels or even the surface. On the basis of this conclusion, the Pinatubo monitoring team raised the alert level.

No measurable changes occurred at the quadrilateral array throughout the month of May, and measurements ceased at the end of May, owing to the high hazard of working at the site. Beginning June 4, a change in the locus of volcano-tectonic seismicity from an area 5 kilometers northwest of Pinatubo to the area beneath the summit dome and active fumarolic vents was accompanied by deformation registered by the closest tiltmeter. The tiltmeter detected about 50 microradians of cumulative tilt between June 4 and June 7 that ended when magma presumably reached the surface and a lava dome began to form. No further cumulative deformation was recorded by the tiltmeter before it was destroyed by the third large explosive eruption on June 13. The tiltmeter recorded only deformation caused by the formation of the conduit from the top of the magma body (5 to 7 kilometers deep) to the surface, not the emplacement of the magma body.

Note to readers: Figures open in separate windows. To return to the text, close the figure's window or bring the text window to the front.


Deformation measurements made prior to the climactic 1991 eruption sequence at Mount Pinatubo, although sparse, proved useful in the analysis of preeruptive unrest. These measurements came from two telemetered tiltmeters on or near the volcano and from a small quadrilateral array of points across the chain of craters that opened on April 2.

Following the onset of unrest at Mount Pinatubo on April 2, 1991, the Philippine Institute of Volcanology and Seismology (PHIVOLCS) began monitoring seismic activity in the area northwest of the volcano. The U.S. Geological Survey (USGS) Volcano Crisis Assistance Team (VCAT) responded to a request from PHIVOLCS for monitoring assistance in the third week of April. Uncertain funding arrangements surrounding the VCAT response led to a "staged response" wherein a three-man team, a portable seismic monitoring system, and a correlation spectrometer (COSPEC) were dispatched to the Philippines to assist PHIVOLCS in assessing the volcanic hazards at Pinatubo and to determine further monitoring needs. Measurements of the quadrilateral array began on May 1, with three subsequent measurements taken that month. By mid-May a telemetered nine-component seismic net had been installed with the central recording and analysis site on Clark Air Base. Constantly increasing seismic activity, a high SO2 emission rate, and a better understanding of the magnitude of previous eruptions convinced us that a more extensive monitoring effort was warranted. By the third week of May additional equipment and personnel arrived at Pinatubo to undertake work that included deformation monitoring.

Two telemetered tiltmeters were installed during the last week of May, and one proved useful in interpreting activity prior to the climactic eruptive phase. Data from the installation nearest the summit tracked the formation of the magma conduit from which the lava dome was extruded on June 7 and showed interesting responses at the beginning of the climactic eruption phase. Various instrumental and logistical problems, along with the increasing volcanic hazard, prevented us from getting many useful data out of the second tiltmeter.


Telemetered tiltmeters are the first deformation monitors VCAT installs in a crisis response because they provide near-real time, microradian-resolution data, they operate in all weather conditions, and can be installed in high-hazard areas that may quickly become too dangerous to revisit. Telemetered deformation data provide a nearly continuous record that allows volcanologists to track changes at a volcano that may occur in a shorter timeframe (minutes to hours) than can be reliably monitored by use of more standard geodetic monitoring techniques such as electronic distance meters or leveling measurements.

VCAT uses bubble-type, biaxial platform tiltmeters with a resolution of 0.1 rad, and a digital data telemetry system designed at the Cascades Volcano Observatory (CVO) (Murray, 1988, 1992a). Telemetered data are received and processed every 10 min at a central monitoring site, which in this case is at Clark Air Base, by use of software developed at CVO that permits the tilt data to be compared easily and rapidly to other monitoring data (Murray, 1990a,b, 1992b).

We transmit four quantities from the tilt site: two perpendicular tilt axes (radial tilt and tangential tilt), temperature at the tiltmeter, and battery voltage. The temperature measurement allows us to look for temperature effects, especially diurnal variations in the tilt values. Tiltmeters are sensitive to even slight changes in temperature, so burial 1 to 2 m deep is necessary in order to isolate the instrument from diurnal temperature changes (Dzurisin, 1992). The battery-voltage measurement allows us to look for supply-voltage effects on the data quality.

To simplify logistics, tiltmeters were colocated with seismometers at the PPO and UBO sites (fig. 1).

Figure 1. Telemetered tilt and seismic stations in place before the June 15, 1991 eruption of Mount Pinatubo. Altitude in meters.


The UBO site was at approximately 1,200 m altitude on the east side of the summit dome, approximately 1,000 m southeast of the active fumaroles, on an abandoned geothermal well pad (fig. 2). The tiltmeter installation at UBO is a typical VCAT tilt site. This style of tilt installation requires about two days to construct (fig. 3).

Figure 2. The UBO site, an abandoned geothermal well pad on the east side of Mount Pinatubo summit dome.

Figure 3. Schematic of the UBO tiltmeter installation on the east side of the Mount Pinatubo summit dome.

To prepare the site we dug a hole approximately 1.5 m deep, which passed through the well-pad fill into the undisturbed soil below. After driving half-inch reinforcing rod (rebar) about 2 m long through the bottom of the hole, we poured concrete around the exposed rebar ends and leveled it. The next day, after the concrete had set, the instrument was placed in the center of the pad and electronically leveled, the top half of a 55-gallon drum was placed over it, and the hole was backfilled. Line-of-sight telemetry was established to Clark Air Base and the first data were received on May 31.


The PPO seismic station was installed at the Aeta village of Patal Pinto, 4 km north of Pinatubo, on May 27 (fig. 1). The instrument was installed in an unused guard house, placed there by the U.S. Air Force for the nearby Crow Valley bombing range. Time constraints and the likelihood of vandalism prevented us from making a standard tiltmeter installation, as at UBO. The tiltmeter was placed on the concrete floor of the guardhouse in hopes of avoiding the need for the laborious standard installation process. The tilt signal was summed onto the PPO seismic telemetry and transmitted to Clark Air Base via a microwave telephone link from the Crow Valley bombing range (Lockhart and others, this volume). The tilt station was installed on May 27, but problems encountered in the field and at Clark Air Base prohibited us from receiving data from the tiltmeter until June 6. Owing to the short time span and active state of the volcano during its operation, we cannot be sure whether reliable data might have been obtained from this type of installation. Figure 4 shows diurnal temperature fluctuations of about 2°C, and diurnal battery-voltage fluctuations of about 0.2 V (owing to the use of a solar panel to charge the station batteries). As these fluctuations are not expressed in any regular manner on the tilt signals, the apparent changes might be real. However, we cannot interpret the PPO tiltmeter data, because the record is too short. As interesting as these tiltmeter data may be, they are probably scientifically useless without an adequate baseline to compare changes against. On the morning of June 10, shortly after the Clark Air Base personnel were evacuated from the Crow Valley installation, the microwave telephone link crucial to reception of the north side of the telemetry net (including the PPO tiltmeter) was vandalized, and no further data were received.

Figure 4. Entire data set from the PPO tiltmeter, north side of Mount Pinatubo. Diurnal changes in the battery voltage (due to the solar panel) and temperature can be seen. Radial and tangential are two perpendicular tilt axes.


On May 1, the USGS-PHIVOLCS team established a quadrilateral array of points across the rift that included the chain of craters created in the April 2 explosive event (fig. 5). Access to this area was by helicopter only. The measurement points consisted of meter-long pieces of rebar driven into the April 2 tephra; only the last centimeter of the rebar remained above the surface. A fifth rebar was driven in about 35 m away from the rift to provide a check on the stability of the marks near the rift (fig. 5). The distance between points across the rift was approximately 20 m. Measurements were made with a steel tape measure held by hand on the tops of the rebars. Typical precision for taping measurements on level ground is +-1/5,000 (Davis and others, 1981). On the basis of this figure, measurement precision of +-4 mm might have been possible. However, owing to time constraints on the use of the helicopter, there was insufficient ground time to make repeated measurements, and there was no means available to apply uniform tension each time the array was measured. Measurement precision was estimated at +-10 mm (+-1/2,000) by the field crew (C.G. Newhall, written commun., 1993).

Figure 5. The rift created on the north-northeast side of Mount Pinatubo in the April 2, 1991, explosion. Letters indicate rift-monitoring station measurement points. View is from the north. The preeruption summit is out of view, to the upper right.

Four sets of measurements were made (table 1), the last on May 28.

Table 1. Data from rift-monitoring station, north-northeast side of Mount Pinatubo.

[Line designations refer to figure 5; measurements in meters]


May 1

May 11

May 16

May 28

Cumulative Difference
















































- 0.003

If movement on the rift occurred between May 1 and May 28, it was too small to be measured accurately with this method. Accelerating unrest and increasing demands placed on the monitoring group's time prohibited further measurements of this array.


May 31-June 3

The radial axis of the UBO tiltmeter was oriented nearly east-west (N.86°W.) such that increasing values would indicate west up, or edifice inflation. The tangential axis was oriented nearly north-south (N.4°E.), such that increasing values would indicate south up. Starting almost immediately after installation, the UBO tiltmeter record was reasonably stable. No diurnal changes were seen in the temperature or tilt data. The first 3 days of operation showed a stable radial component and a slight north-up trend on the tangential component (fig. 6).

Figure 6. Data from the UBO tiltmeter and seismometer on the east side of the Mount Pinatubo summit dome. Rainstorm on June 3 is indicated, as is the ash emission that followed. RSAM, Real-time Seismic Amplitude Measurement.

These steady trends were interrupted on June 3 when a heavy rainstorm came through the area at about 1400 local time, and apparently caused significant tilt and temperature excursions (fig. 6). Two explanations of these excursions are reasonable: a local site response to the rainstorm or an instrumental or electronic effect in which moisture from the rain affected the equipment. We feel that the tilt and temperature excursions are probably due to local site effects and are not instrumental or electronic. Rapid changes in tilt after heavy rainstorms have also been observed at Mount St. Helens (Dzurisin, 1992).

Earlier in the day (1130 local time) we had changed the batteries at the UBO tilt installation. The telemetered battery reading shows the jump in voltage. The radial and tangential axes show minor stepwise changes at this time, possibly resulting from site settling caused by our activity around on the installation. If settling occurred, then the heavy rain a few hours later could readily have caused additional settling of the site.

Later on June 3, a small ash emission occurred that was accompanied by a high-frequency (4-6 Hz) tremor signal on the UBO seismic station. A small transient inflationary spike on the radial axis and a smaller north-down spike on the tangential axis coincide with this event. Soon after the small explosion, the locus of volcano-tectonic earthquakes shifted from about 5 km northwest of Pinatubo to the area immediately beneath the summit dome (Harlow and others, this volume). At this time the tangential tilt component returned to its previous slow, north-up trend, and the radial component displayed a more decided inflationary trend.

June 4-June 7

The gradual inflationary tilt trend on the radial axis continued until about 1200 local time on June 6. Then it began to accelerate, coincident with intensifying shallow seismicity beneath the summit dome (fig. 7). Over the next 30 h more than 2,000 volcano-tectonic earthquakes occurred 0 to 3 km beneath the summit dome. In the last 4 h of the seismic swarm, the tilt signal on both axes became increasingly noisy, but the tilt trend remained clear; this indicates that the tilt signal was noisy owing to seismic ground shaking, not to electronic problems. At about 1700 local time, a sudden gas and ash emission sent a plume to about 4,000 m. Immediately thereafter, the radial tilt signal leveled off but remained noisy for about another 8 h.

Figure 7. Radial and tangential components of the UBO tiltmeter compared to UBO seismometer Real-time Seismic Amplitude Measurement (RSAM) during days preceding dome extrusion, east side of the Mount Pinatubo summit dome.

The tangential tilt axis at UBO continued tilting toward the south at a constant rate until about 2000 local time on June 6, when the tilt rate accelerated slightly, lagging the response of the radial axis by some 8 h. During this period, the tangential axis showed a response similar to that of the radial axis but did not mirror it, as might have been the case if electrical crosstalk had occurred between the electronics of the two axes. In addition, neither the battery-voltage nor temperature channels showed changes mirroring those seen on the tilt axes. Thus, we feel confident that the June 6-7 tilt excursion of the radial tilt axis reflects a real tilt away from the summit area and, possibly, a lesser tilt toward the south.

We interpret the increasing tilt from late on June 3 through the occurrence of the sudden steam and ash emission on June 7 as marking the formation of the magma conduit to the surface, and we interpret the abrupt flattening of the tilt signal as marking the time when magma broke through to the surface (see figures 5-8 in Harlow and others, this volume). At the time these events occurred we interpreted the combined seismic and tilt data to mean that, at the very least, magma had reached very shallow levels or even the surface. On the basis this conclusion an Alert Level 4 was declared (Ewert and others, 1991) (see Punongbayan and others, this volume, for a description of the alert level system). Visual observations on the morning of June 8 confirmed the presence of a new lava dome.

June 8-June 13

After the dome extrusion, ash emission became more voluminous and episodes of tremor more frequent and of longer duration, but no more cumulative tilt events occurred (fig. 8). Transient tilt events continued until the first large eruption on June 12. These events were about 10 rad in amplitude and were associated with short-lived swarms of high- frequency earthquakes but not with tremor episodes. Tilt transients are also associated with explosions at 0341 and 0851 local time on June 12 (fig. 8). The tilt signal returned to baseline at the end of each swarm.

Figure 8. Last 5 days of operation of the UBO tiltmeter with Real-time Seismic Amplitude Measurement (RSAM) data from UBO and PIE (fig.1) seismometers, east side of Mount Pinatubo. Note generally flat tilt from June 8 through June 11 compared to RSAM data on same days. Peaks in RSAM data correspond to periods of tremor.

In the June 8-12 time period, the character of seismicity changed from being dominated by volcano-tectonic events and periods of high-frequency (3-6 Hz) tremor to dominantly long-period seismicity with periods of low-frequency tremor (<1 Hz). We see no correlation of tilt events with long-period swarms or tremor during this time period.


Beginning on June 4, a change in the locus of volcano-tectonic seismicity from 5 km northwest of Pinatubo to beneath the summit dome and active vents accompanied deformation registered by the UBO tiltmeter. This shift in seismicity probably marked the formation of the magma conduit to the surface (Harlow and others, this volume). The tiltmeter detected about 50 rad of cumulative tilt between June 4 and June 7, ending when magma presumably reached the surface and a lava dome began to form. From June 7 to June 13, the tilt signal from UBO was noisy during periods of increased seismicity, probably due to ground shaking, but the signal always returned to its background level.

Measurements across the April 2 rift were confined to the month of May, and no deformation signal was apparent. We had planned to install a tiltmeter at this site during the first week of June, but the rapidly evolving unrest and other priorities associated with the response ultimately prevented the installation. Both the UBO tilt/seismic site and the rift-monitoring station were very close to what would become the caldera margin on June 15.

No signs of gross deformation (ground cracking or faulting) were observed in the rift area or the summit region during helicopter inspections between June 7 and 14. These observations, coupled with the small tilt signal observed from the UBO tiltmeter, may seem surprising given the large volume of erupted material (3.7-5.3 km3 of dense rock equivalent) (W.E. Scott and others, this volume) and the even larger inferred volume (40-90 km3) of the magma chamber (Mori, Eberhart-Phillips, and Harlow, this volume). However, Mount Pinatubo has been the site of recurring large silicic eruptions throughout the Holocene, the most recent only 600 years ago. (Newhall and others, this volume). Smith (1979) showed that silicic magma chambers erupt only about 10 percent of their volume during any one eruptive period, and Pallister and others (this volume) show that the trigger for the 1991 eruption of Mount Pinatubo was an injection of basaltic magma into a preexisting dacitic magma chamber. Therefore, most of the magma volume was probably already in place long before monitoring began.

From the small tilt changes and the lack of any obvious ground deformation prior to the Plinian phase of the eruption, we infer that no cryptodome was emplaced near the surface and that the eruptions proceeded from the overpressurized magma chamber, the top of which was about 5 to 7 km deep (Mori, Eberhart-Phillips, and Harlow, this volume; Pallister and others this volume). Thus, the tilt changes recorded only the formation of the conduit from the magma chamber to the surface and did not record the accumulation of magma beneath the summit.


We would like to thank Art Daag (PHIVOLCS), Bella Tubianosa (PHIVOLCS), Chris Newhall (USGS), and Rick Hoblitt (USGS) for making the measurements at the rift station. Thanks are also due to the helicopter pilots and flight crews of the U.S. Thirteenth Air Force, Third Tactical Fighter Wing, for excellent logistical support in trying circumstances, and to Don Swanson and Dan Dzurisin for helpful comments during the preparation of this manuscript.


Davis, R.E., Foote, F.S., Anderson, J.M., and Mikhail, E.M., 1981, Surveying theory and practice: New York, McGraw-Hill, 992 p.

Dzurisin, D., 1992, Electronic tiltmeters for volcano monitoring: Lessons from Mount St. Helens, in Ewert, J.W., and Swanson, D.A., eds., Monitoring volcanoes: Techniques and strategies used by the staff of the Cascades Volcano Observatory, 1980-90: U.S. Geological Survey Bulletin 1966, p. 69-83.

Ewert, J.W., Lockhart, A.B., Hoblitt, R.P., and Harlow, D.H., 1991, Ground tilt events and the critical June 5-7, 1991, precursory volcanic activity at Mt. Pinatubo, Philippines [abs.], Eos Transactions, American Geophysical Union, v. 72, no. 44, p. 61.

Harlow, D.H., Power, J.A., Laguerta, E.P., Ambubuyog, G., White, R.A., and Hoblitt, R.P., this volume, Precursory seismicity and forecasting of the June 15, 1991, eruption of Mount Pinatubo.

Lockhart, A.B., Marcial, S., Ambubuyog, G., Laguerta, E.P., and Power, J.A., this volume, Installation, operation, and technical specifications of the first Mount Pinatubo telemetered seismic network.

Mori, J., Eberhart-Phillips, D., and Harlow, D.H., this volume, Three-dimensional velocity structure at Mount Pinatubo, Philippines: Resolving magma bodies and earthquakes hypocenters.

Murray, T.L., 1988, A system for telemetering low-frequency data from active volcanoes: U.S. Geological Survey Open-File Report 88-201, 28 p.

------1990a, A user's guide to the PC-based time-series data-management and plotting program BOB: U.S. Geological Survey Open-File Report 90-56, 53 p.

------1990b, An installation guide to the PC-based time-series data-management and plotting program BOB: U.S. Geological Survey Open-File Report 90-634-A, 25 p.

------1992a, A low-data-rate digital telemetry system, in Ewert, J.W., and Swanson, D.A., eds., Monitoring volcanoes: Techniques and strategies used by the staff of the Cascades Volcano Observatory, 1980-90: U.S. Geological Survey Bulletin 1966, p. 11-23.

------1992b, A system for acquiring, storing, and analyzing low-frequency time-series data in near-real time, in Ewert, J.W., and Swanson, D.A., eds., Monitoring volcanoes: Techniques and strategies used by the staff of the Cascades Volcano Observatory, 1980-90: U.S. Geological Survey Bulletin 1966, p. 37-43.

Newhall, C.G., Daag, A.S., Delfin, F.G., Jr., Hoblitt, R.P., McGeehin, J., Pallister, J.S., Regalado, M.T.M., Rubin, M., Tamayo, R.A., Jr., Tubianosa, B., and Umbal, J.V., this volume, Eruptive history of Mount Pinatubo.

Pallister, J.S., Hoblitt, R.P., Meeker, G.P., Knight, R.J., and Siems, D.F., this volume, Magma mixing at Mount Pinatubo: Petrographic and chemical evidence from the 1991 deposits.

Punongbayan, R.S., Newhall, C.G., Bautista, M.L.P., Garcia, D., Harlow, D.H., Hoblitt, R.P., Sabit, J.P., and Solidum, R.U., this volume, Eruption hazard assessments and warnings.

Scott, W.E., Hoblitt, R.P., Torres, R.C., Self, S, Martinez, M.L., and Nillos, T., Jr., this volume, Pyroclastic flows of the June 15, 1991, climactic eruption of Mount Pinatubo.

Smith, R.L., 1979, Ash-flow magmatism, in Chapin, C.E. and Elston, W.E., eds., Ash-flow tuffs: Geological Society of America Special Paper 180, p. 5-25.

FIRE and MUD Contents

PHIVOLCS | University of Washington Press | U.S.Geological Survey

This page is <>
Contact: Chris Newhall
Last updated 06.10.99