Smithsonian Institution
Global Volcanism Network Bulletin v. 20, no. 9, September 1995

Ruapehu (New Zealand)  Large eruptions produce lahars and send
       plumes to over 10 km altitude

North Island, New Zealand
39.28S, 175.57E; summit elev. 2,797 m
All times are local (= GMT + 12 hours)

Following noteworthy "vent clearing" eruptions at Ruapehu (figure 1)
on 29 June and 3 July, and phreatic eruptions in September, a
series of larger eruptions began on 23 September. During the next
week Ruapehu discharged plumes that were frequently reported by
aviation sources to have reached at least 10 km. The following was
compiled from Institute of Geological & Nuclear Sciences (IGNS)
reports and aviation notices.

Precursory activity and minor eruptions. Many of Ruapehu's frequent
small eruptions have been linked to high temperature in the crater
lake. Unusually high lake temperatures (as well as other measured
changes) also preceded the recent activity. During 1985-95 the
surface temperature of Ruapehu's crater lake peaked at >40 deg C
seven times; two of those peaks were in 1995. The early 1995 peak
reached 55 deg C, the highest surface lake temperature recorded in
13 years (Bulletin v. 20, nos. 1 and 5). The second 1995 peak
reached roughly 44 deg C, the third highest seen in the 1985-95
interval. Key observations, including those from crater lake
inspections carried out during visits from 25 May through 23
September (table 1) suggested a build-up in activity.

A hydrophone and related acoustical detection components in the
crater lake registered unusually high noise levels during late May,
consistent with seismic activity. A moderate noise burst took place
on 13 June, and relative quiet prevailed through 29 June. These
data were communicated via the satellite-relayed ARGOS data system
at 2-hour intervals; the last transmission (0800 on 29 June) came
just prior to a M 3.2 volcanic earthquake and eruption that
destroyed the ARGOS equipment.

Seismicity was at background levels from 15 May until just prior to
the 29 June earthquake. The earthquake began at 0802 as a small 2-
Hz signal followed by a 1-Hz signal. The main part of the
earthquake, which also contained 2-Hz signal, started at 0821, and
peaked between 0822 and 0824. After the main part of the
earthquake, more signals centered around 1 and 2 Hz prevailed. The
2-Hz signals are common to both volcanic earthquakes and tremor at
Ruapehu, suggesting that both may excite the same resonator.

Ruapehu's tremor typically has a dominant frequency of ~2 Hz and
occurs almost constantly, often with no clear surface volcanic
expression. Although not recorded at all stations, during 1995 and
possibly longer, tremor has contained a previously unrecognized
dominant frequency of 7 Hz with a consistent amplitude of 1
fm/second. During April, May, and late June, intervals of strong 2-
Hz tremor dominated the seismic records. Very strong tremor took
place for a few hours on 26 June. Tremor declined thereafter and
remained low from early July through much of August.

Lake water increased in Cl and especially Mg ions closer to the
eruption. The Mg/Cl ratio rose from values around 0.035 in early
1995 (Bulletin v. 20, no. 5), to the most recently reported value
of 0.072 on 15 August (table 2); there was a further increase of
unstated magnitude on 20-21 September (table 1). Prior to the
eruption, the rise in Mg was thought to represent leaching from
unweathered andesites. The increase in Cl, which reached greater
levels than seen in at least 9 years, was thought to result from
both large-scale evaporation and HCl input. The rise in Mg/Cl ratio
represented the largest shifts seen since the large 1971 and 1975
eruptions. Shifts in the concentrations of K, Fe, and SO4 from
samples collected on 18 July suggested increased input of SO2 into
the vent-lake system rather than a water-rock equilibrium process
in the vent. Although provisional, results for SO4 on 18 July
suggested a 4.5% increase--the highest ever recorded for the lake.

Deformation surveys on 4 July and 15 August confirmed only small
measurable changes. This result suggested little or no magmatic
movement in the upper part of the vent, in contrast with much of
the other data in the same time interval. The limited deformation
may have been a consequence of an open vent that allowed a small
amount of magma to escape without measurable deformation.
Measurable changes were apparently evident later (20-21 September,table 1).

Larger eruptions in late September. Ruapehu produced a series of
larger eruptions during 23-30 September and later, continuing into
October. Preliminary estimates suggested the eruption plumes
reached 8-12 km heights as reported by aviation sources (table 1
and figure 2). The aviation reports and occasional satellite
imagery typically noted plumes possibly extending as far as ~270 km
from the summit (from an episode of eruptive bursts that were
thought to have been more dense and ash-rich beginning at 1600 on
24 September). This particular series of bursts only initially
reached low levels, but ash was said to have been lifted higher by
induced cumulus convection, ultimately reaching a reported altitude
of ~12 km. On subsequent days, the plume's typical maximum lateral
extent was given as roughly 60 km.

For the 24-hour interval ending on 24 September (exact times
undisclosed) observers at Ruapehu noted both small- and medium-
sized steam-rich ash-bearing explosions, the largest of which had
plumes that rose from 500 to over 1,500 m. On 24 September medium-
sized explosions yielded a distinctive, though modest seismic
signature and lesser explosions were not detectible. Near midnight
on 24 September the number of volcanic earthquakes rose
significantly; strong tremor roughly doubled in intensity compared
to that morning; reflected seismic waves from numerous explosions
yielded a confused signal.

Reports for 25 September (at 0900, 1700, and a summary the next
day) noted that an eruption column had developed from many
moderate-sized eruptions. With its top at 8-10 km altitude, the
plume was blown into the E quadrant for  several tens of
kilometers, dropping ash 18 km E (Desert road; total accumulation,
1 mm), 30 km E (the Kaimanawa mountains), and 120 km E (traces at
the coast). The ash deposited at Desert Road contained mainly
particles of 10-250 fm size; 30-60% of the particles were juvenile.
Significant amounts of ash had accumulated in the vent area but
large blocks had been ejected less than ~1 km from the vent. Outlet
was dry, but based on later observations, the inner crater still
contained a lake.

At 0900 on 25 September a lahar flowed down the Whangaehu valley.
The valley forms a key drainage that descends ESE from the crater,
ultimately curving S and W to encircle Ruapehu's S flank;
downstream parts of the Whangaehu Valley cross the Auckland-
Wellington rail line near Tangiwai. Later the lahar declined in
size, but it was noted as still continuing and sediment-laden at
1630, having eroded a stream bank upstream of the Tangiwai bridge.
Another lahar flowed W of the crater down Mangaturuturu Valley.

At 1700 on 25 September, the volcanism during the previous 30 hours
was described as episodic, punctuated by two cycles of increasing
then decreasing intensity. Based on seismic data, the second cycle
was not quite as vigorous as the first. In the night and morning of
25-26 September minor amounts of ash continued to fall over the
volcano's E quadrant. Low-to-moderate tremor continued until at
least 1700. Occasional explosions were large enough to be recorded
seismically but were smaller than those in the morning of the
previous day. Although during much of the day visual observations
were hampered by cloud cover, at 0600-0700 on 26 September
observers saw the plume drifting ESE. The plume was fed by numerous
weak explosions and observers noted that minor amounts of ash fell
throughout the night. Observers also noted that lahars flowing down
the Whangaehu Valley were smaller than on the previous day. A very
small lahar, deposited during an earlier event, was noted in the
SE-flank Wahionoa Valley.

A SO2 flux measurement at 1600 on 26 September indicated an output
of 2,600 +- 400 metric tons/day. Such high fluxes confirmed
significant magmatic involvement in the eruption. Although cloud
cover limited the visibility on much of 26 September, the low
seismic activity during the day suggested explosions of modest
size. From about 2300 through early the next morning tremor
amplitude fluctuated, increasing up to moderate levels. After 0400
tremor coexisted with many volcanic earthquakes.

Visual observations made after sunrise on 27 September correlated
tremor and earthquake increases to moderately vigorous eruptive
activity. During this period (0600-0700) the earthquakes reached a
size equivalent to those on 25 September. By about 0930 on 27
September, however, the earthquakes stopped and the eruption's size
dropped. Earthquakes then remained undetected until at least 1700.

Aerial observers on 27 and 28 September saw that Crater Lake had
been greatly reduced in size; although indistinct, the steaming
surface had clearly dropped by tens of meters. They  also saw a
previously concealed terrace formed during the 1945 eruption and
recognized a new small lahar deposit in a drainage on the NW flank
(in the Whakapapaiti Valley). On 27 September observers reported no
water in the upper Whangaehu Valley and viewers the next day stated
that downstream at the Tangiwai bridge the water level had returned
to normal.

During the 24 hours ending at 0930 on 28 September, moderate levels
of seismicity prevailed, and three larger volcanic earthquakes took
place in the 0215-0340 interval. These earthquakes may have been
associated with discrete explosions. Other volcanic earthquakes at
0736 and 0839 were linked to mild puffs of ash-bearing steam rising
from the crater.

Ruapehu's alert status was raised to Level 4 (table 3) on 25
September. As late as early October, there had been no reports of
death or injury caused by the eruption. Because of potential hazard
to aircraft, aviation and meteorological workers have carefully
monitored the eruption, producing forecasts of the plume's
transport and dispersal ("VAFTAD" modeling program, see Bulletin v.
19, no. 6) as well as the actual visible observations that have
confirmed the height of the plume's top (figure 2).

The late September eruption was widely covered in the news.
According to Reuters (25 September), "A conservative Australian
politician is linking nuclear testing by China and France to a
string of earthquakes around the Pacific and volcanic eruptions in
Montserrat and New Zealand's Mount Ruapehu." Although this
connection was discounted by earth scientists, the accusation did
reverberate in the media and parliaments world wide.

Ruapehu is a complex stratovolcano with five Holocene vents on the
summit and flanks. Crater Lake is currently Ruapehu's sole active
vent. The lake, which is frequently the site of geyser-like gushing
ejections of hot water, lies cradled in a broad summit depression.
The valleys draining the volcano have been inundated by repeated
lahars. The volcano's summit is glacially covered and an alpine ski
area lies on its N flanks.

Information Contacts: C.J.N. Wilson, B.J. Scott, P.M. Otway, and
I.A. Nairn, Institute of Geological & Nuclear Sciences (IGNS),
Private Bag 2000, Wairakei, New Zealand (Email:;
Bureau of Meteorology, Northern Territory Regional Office, POB 735,
Darwin NT 0801, Australia J. Heffter, National Oceanic and
Atmospheric Administration (NOAA), Air Resources Laboratory SSMC3,
Room 3151, 1315 East West Hwy., Silver Spring, MD 20910 USA (Email:
nick@arlrisc.ssmc.; Synoptic Analysis Branch,
NOAA/NESDIS, Room 401, 5200 Auth Road, Camp Springs, MD 20746 USA.

Figure 1. Index map of North Island, New Zealand, showing the
location of Ruapehu and other volcanic centers.

Figure 2. Histogram summarizing the height of column tops for
Ruapehu eruptions, based on available aviation reports and IGNS
Science Alert Bulletins. Courtesy of Nick Heffter, NOAA.

Table 1. Summary of key observations at Ruapehu, 25 May-23
September 1995. Prior to the larger eruptions observers reported
that the lake was generally gray in color, often with sulfur slicks
on its shore or surface; the lake began discharging water at Outlet
sometime between 4 and 18 July. Courtesy of IGNS.

Date    Crater Lake Data  Other Observations (2nd line)

25 May  45.9 deg C at Outlet; ~0.7 m below overflow.
        One very small eruption observed.
16 Jun  38.0 deg C at Outlet; ~1.5 m below overflow.
        No evidence of recent eruptions.
26 Jun  Very strong tremor for a few hours at one station.
29 Jun  Last ARGOS transmission.
        Volcanic earthquake (M 3.2) correlated with an eruption.
03 Jul  --
        Volcanic earthquake (M 2.4) correlated with an eruption.
04 Jul  33.0 deg C at Outlet; 0.5 m below overflow.
        Intense steaming in the lake center. Two very small
        eruptions observed; evidence of larger eruptions that
        probably occurred on 29 June and 3 July. Small deformation.
18 Jul  31.0 deg C at Outlet. Discharge of 50 l/s.
        Evidence for recent minor eruptions but no observed
15 Aug  29.0 deg C at Outlet. Discharge of 5-10 l/s.
        No evidence of recent activity; small deformation.
18 Sep  Moderate vent-clearing explosive eruption at 0805 from
        within the lake.
        Caused a flood, a lahar, and a small mudflow down the
        flanks; accompanying volcanic earthquake (ML 3.6). The
        lahar was the largest down the Whangahu river since 1975.
20 Sep  48 deg C at Outlet. Very large overflow.
        New scoria bombs found; 15 small phreatic eruptions
20 Sep  --
        Eruption similar to 18 September,, only smaller;
        accompanying volcanic earthquake (ML 3.2).
20-21 Sep  Lake water chemistry indicates increased magma-water
           Geodetic data show increased crater diameter.
23 Sep  Major eruption began; column top reached over 10 km

Sources: IGNS Immediate Report (25 May - 15 Aug); IGNS Science
Alert Bulletin (18-21 Sep); Aviation report (23 Sep).

Table 2. Ruapehu Crater Lake water analyses and temperatures at
Outlet, 25 May-4 July 1995. Courtesy of IGNS.

Date       Mg (ppm)   Cl (ppm)   Mg/Cl  Temp  deg C

25 May 95    385       7,603     0.051    46
16 Jun 95    427       7,797     0.055    38
04 Jul 95    514       7,976     0.064    33
18 Jul 95    551       8,014     0.069     --
15 Aug 95    584       8,154     0.072     --

Table 3. Scientific Volcano Alert Level system for New Zealand
volcanoes. Courtesy of the IGNS.

Level  Phenomena Observed/Scientific Interpretation (Volc Status)

0    Typical background surface activity; seismicity, deformation,
     and heat flow at low levels. Usual dormant, intra-eruption or
     quiescent state.

1    Departure from typical background surface activity. Minor
     phreatic activity.
     Apparent seismic, geodetic, thermal, or other unrest
     indicators. Signs of volcano unrest. No significant eruption

2    Increase from a low level of activity, accompanied by changes
     to monitored indicators. Significant change in level or style
     of ongoing eruptive activity.
     Increase in seismicity, deformation, heat flow and/or other
     unrest indicators.  Indications of intrusive processes. Local
     eruption threat.

3    Increased vigour of ongoing activity and monitored indicators.
     Significant local eruption in progress.
     Commencement of minor eruptions at reawakening vent(s).
     Relatively high and increasing trends shown by unrest
     indicators. Increasing intrusive trends indicate real
     possibility of hazardous eruptions.

4    Significant change to ongoing activity and monitored
     indicators. Hazardous local eruption in progress.
     Establishment of magmatic activity at reawakening vent(s),
     with acceleration of unrest indicators. Large-scale eruption
     now appears imminent.

5    Hazardous large volcanic eruption in progress. Destruction
     within the Permanent Danger (red) Zone. Significant risk over
     wider areas.

Note that the frequently active cone volcanoes of New Zealand
(White, Ngauruhoe, and Ruapehu) require definitions different from
all other volcanic systems. Because of this, Alert Levels 1-4 are
split into two parts: one for the frequently active cones and the
other for reawakening systems. Comments on this approach would be
appreciated, and should be sent to Brad Scott, IGNS.