Bull Volcanol (2010) 72:279–296 DOI 10.1007/s00445-009-0318-2
RESEARCH ARTICLE
Depositional record of historic lahars in the upper Whangaehu Valley, Mt. Ruapehu, New Zealand: implications for trigger mechanisms, flow dynamics and lahar hazards Alison H. Graettinger & Vern Manville & Roger M. Briggs
Received: 14 October 2008 / Accepted: 24 August 2009 / Published online: 11 September 2009 # Springer-Verlag 2009
towards either the largest or the most recent flows. In some cases magnitude can be reconstructed from deposit geometry, with the largest lahars producing the highest level terraces, the coarsest deposits, and crossing drainage divides into normally inactive channels. This under-representation of historic events reflects the low preservation potential of unconsolidated deposits in a steep alpine environment, and the overprinting and recycling effect of large magnitude lahars that rework material down to bedrock and effectively reset the stratigraphic record. Development of magnitudefrequency relationships for Ruapehu lahars therefore requires the identification of lahar deposits in proximal, medial and distal settings in order to ensure that the full range of events is represented.
Abstract Mt. Ruapehu, in the central North Island of New Zealand, is one of the most lahar-prone volcanoes in the world. Since historic observations began in 1861 AD, more than 50 individual lahars have been recorded in the Whangaehu valley alone, the natural outlet to the summit Crater Lake. These lahars have been triggered by a variety of mechanisms, including explosive eruptions that displaced Crater Lake water over the outlet or ejected it onto the snow-clad summit area of the volcano; rainremobilisation of tephra deposits on steep slopes; displacement over the outlet as a result of syn-eruptive changes in lake bathymetry; and lake break-outs from Crater Lake following impoundment of excess water behind temporary barriers of tephra and/or ice emplaced over the outlet. However, only 9 lahar deposits can be distinguished in the upper Whangaehu valley on sedimentological, stratigraphic, geomorphic and petrological grounds, and these are skewed
Keywords Lahars . Ruapehu . Crater lake . Debris flows . Volcanic hazards . Trigger mechanism
Editorial responsibility: J. White
Introduction
A. H. Graettinger (*) : R. M. Briggs Department of Earth and Ocean Sciences, The University of Waikato, Private Bag 3105, Hamilton, New Zealand e-mail:
[email protected] V. Manville GNS Science, Wairakei Research Centre, Private Bag 2000, Taupo, New Zealand Present Address: A. H. Graettinger Department of Geology and Planetary Science, The University of Pittsburgh, SRCC, Room 200, Pittsburgh, PA 15260-3332, USA
The dynamics of lahars, volcanogenic floods, and mass flows are complex functions of their sediment: water ratios, granulometry, temperature, flow velocity, clay mineral content and channel geometry (Pierson and Costa 1987; Costa 1988; Iverson 1997). Typical lahars on Mt. Ruapehu evolve along their flowpaths, transforming reversibly and repeatably, from initially clear-water Newtonian flows to hyperconcentrated flows to non-cohesive debris flows, due to the interplay between bulking by entrainment of particulate matter and dilution through deposition or incorporation of ambient stream water (Pierson and Scott 1985; Scott 1988). Snow or ice slurry lahars are a distinct category whose behaviour is dominated by incorporated particulate frozen water (Major and Newhall 1989; Pierson
280
and Janda 1994; Waitt et al. 1994; Cronin et al. 1996; Lube et al. 2009). Globally, lahars are the third deadliest volcanic hazard historically (Tanguy et al. 1998; Witham 2005): New Zealand’s own worst volcanic disaster was a break-out lahar from Ruapehu’s Crater Lake in 1953, 8 years after an eruption created a temporary dam at the outlet (O’Shea 1954; Manville 2004). Mitigation of future lahar hazard at Ruapehu and other crater lake- or glacier-bearing volcanoes relies on an understanding of lahar processes, as guided by studies of active flows (e.g. Manville and Cronin 2007) and quantification of the magnitude and frequency of lahars derived from the geological record. At Ruapehu, the preserved medial (ring-plain) and distal prehistoric lahar record is relatively well known (Campbell 1973; Palmer 1991; Hodgson 1993; Palmer et al. 1993; Cronin et al. 1997a; Lecointre et al. 2004; Hodgson et al. 2007). However, deposits of only four historic lahars are identified and named as the youngest members of the Onetapu Formation (1861, 1953, 1975, and 1995E/R; Palmer 1991; Cronin et al. 1997a; Hodgson et al. 2007). The most recent historic lahars have been thoroughly studied through direct observation, or analysis of their deposits immediately after emplacement (Cronin et al. 1996, 1997b, 1997c, 1999, 2000; Manville et al. 1998, 2007; Hodgson and Manville 1999), none more so than the March 2007 Crater Lake break-out (Manville and Cronin 2007; Marutani et al. 2007; Carrivick et al. 2008). There is, however a knowledge gap of historic lahars in proximal areas, i.e. between source and c. 10 km downstream, corresponding to incised gorges on the volcanic edifice itself. This paper concentrates on this area of the Whangaehu gorge on the eastern side of Ruapehu, between Crater Lake and the apex of the Whangaehu fan (Fig. 1). The Whangaehu valley not only is the most active lahar path on the volcano; the incised nature of the gorge has allowed the short-term accumulation of dynamic diamicton terraces.
Geological setting Mt. Ruapehu is the largest and most active andesitic stratovolcano in the central North Island of New Zealand (Fig. 1), and lies at the southern end of the Taupo Volcanic Zone. The 110 km3 composite cone rises 2,797 m above sea level and supports a number of small glaciers, permanent snow fields and a summit crater lake. It is surrounded by a volumetrically equivalent ring-plain constructed of pyroclastic fall, lahar and fluvial deposits derived from the volcano (Hackett and Houghton 1989). Historic activity at Ruapehu has consisted of very frequent, relatively small-tomedium phreatic and phreatomagmatic eruptions (e.g. 1895, 1969, 1971, 1975, 1978, 1979, 1980, 1981–1982, 1985, 1987,
Bull Volcanol (2010) 72:279–296
1988, 2007) (Gregg 1960; Healy et al. 1978; Nairn et al. 1979; Houghton et al. 1987), and more prolonged magmatic eruptions such as occurred in 1945 (Oliver 1945; Reed 1945; Beck 1950) and 1995–96 (Bryan et al. 1996; Nakagawa et al. 1999). At least five vents have been active on Ruapehu during the Holocene, but historic activity has been confined to the southern crater which is occupied by a hot, acidic Crater Lake, the most recent estimated (pre-1995) volume of which is c. 9 million m3 (Christenson and Wood 1993).
Crater lake and the Whangaehu valley A sector collapse from the eastern flank of Ruapehu at c. 4.6 ka produced the Mangaio Formation debris avalanche (Donoghue et al. 1988; Donoghue and Neall 2001). An ancestral Crater Lake may have developed in the collapse amphitheatre as early as 3 ka (Donoghue et al. 1997), but eruptive activity remained in the Central and North craters between >3 ka and c. 2 ka. Development of a permanent Crater Lake over the southern vent has been correlated with a shift from dry, magmatic strombolian ashes in the lower Tufa Trig Formation to phreatomagmatic/magmatic ash couplets at c. 1.5 ka (Donoghue et al. 1997). This also corresponds to the onset of voluminous lahars of the Onetapu Formation, some deposits of which contain fragments of lake sediment, in the Whangaehu catchment (Palmer et al. 1993; Lecointre et al. 2004; Hodgson et al. 2007). This shift in eruption type may only reflect the initiation of activity in the south crater, under the lake, and as a passive lake may have existed in the collapse crater before eruptive activity of the southern crater commenced. The modern Crater Lake (2,530 m elevation), overflows into the gorge at the head of the Whangaehu valley, which is carved into >60 ka andesite lava flows of the Wahianoa Formation (Hackett and Houghton 1989), modified by glaciation (McArthur and Shepherd 1990), and partially infilled by Holocene lava flows, the 4.6 ka Mangaio Formation, and laharic deposits preserved as discontinuous terraces and barforms (Fig. 2). The valley descends c. 1,300 m in 10 km (distances are measured from the outlet to Crater Lake along the axis of the North Branch active channel) via a steep (avg. 0.01 m/m), tortuous gorge before debouching onto the Whangaehu fan. Water from the lake reaches the valley through two pathways: across the Whangaehu Glacier after ejection from the lake, or overflow at the outlet (Fig. 2). Large primary (eruptiontriggered) and secondary (rain-triggered or lake break-out) lahars can enter the normally dry South Branch of Whangaehu gorge by overtopping drainage divides at 1.5 and 2 km (Fig. 3). The South Branch rejoins the active North Branch c. 500 m upstream of where the Round-the-mountaintrack (RTMT) crosses the river (7.1 km). Large lahars also
Bull Volcanol (2010) 72:279–296
281
Fig. 1 Location map of Mt. Ruapehu and the Whangaehu River
bifurcate around a spur of lava close to the RTMT bridge, and at the head of the Chute (8.4 km) where portions of large flows partially overtop a drainage divide across the back of a (?) Holocene lava flow to enter a southerly-directed confined channel; the remainders of the lahars discharge from the apex
of the fan to enter multiple braided distributary channels. These coalesce back into a single thread channel at the foot of the fan (15–17.4 km) where it abuts the north-striking Whangaehu escarpment of the Rangipo/Desert Road Fault (Donoghue and Neall 2001; Villamor et al. 2007).
282
Bull Volcanol (2010) 72:279–296
Fig. 2 a Oblique aerial photograph of the Whangaehu valley c. 7 km downstream from Crater Lake. The light grey surfaces are the deposits of the 18 March 2007 Crater Lake break-out lahar (image L. Homer, GNS Science). b Oblique aerial photograph of Crater Lake and the head of the Whangaehu gorge (image L. Homer, GNS Science)
Methodology The historic lahar record in the upper Whangaehu valley was reconstructed from field investigations, archival aerial photos, two airborne LiDAR (e.g. French 2003) topographic surveys (commissioned by GNS Science and Massey University) before and after the March 2007 lahar, field and
Fig. 3 Lahar paths of the 1975E lahar in the upper Whangaehu valley and the Whangaehu fan showing principal bifurcation points. Inset shows sample locations (dots): WOB = west of bridge, RTMT = Round-the-mountain-track, EOB = east of bridge, IntEast = intermediate east, and line of section of Fig. 4. The Bund is an earthen
laboratory granulometric and petrographic studies, eyewitness accounts, published reports (Gregg 1960; Healy et al. 1978; Nairn et al. 1979; Otway et al. 1995), and instrumental measurements of active flows. Peak discharge estimates were based on flow cross-sectional areas determined from highwater marks and local velocities derived from the superelevation equation (Johnson and Rodine
structure created by the Department of Conservation to prevent lahars from contaminating the Tongariro River by overtopping a drainage divide, north of the Whangaehu River. New Zealand Map Grid (2000) co-ordinates
Bull Volcanol (2010) 72:279–296
1984), and Manning’s equation (Chow 1959), where the value of Manning’s coefficient, n, was calibrated using the known velocity of historic lahars derived from arrival times and/or the superelevation equation.
Deposit sedimentology and stratigraphy Historic lahar deposits in the upper Whangaehu valley form a series of unconsolidated terraces of very poorly sorted, coarse bouldery gravel and sand. These display surficial gravel and boulder bars (Carling 1989) and streamlined cluster bedforms (Brayshaw 1984). Terraces of older lahars occur in wider portions of the gorge, with narrow segments of the channel containing only the youngest deposits, and in places exposed bedrock. The unconsolidated deposits are easily remobilised and buried by frequent lahar activity and can vary dramatically during eruptive episodes where prolonged activity through the lake can result in lahars every few days (Cronin et al. 1997c). Deposits are predominantly massive, very poorly sorted diamictons with clasts ranging from silt to 5 m diameter boulders, indicative of deposition from non-cohesive hyperconcentrated or debris flows (Pierson and Costa 1987) (Table 1). Such flows cannot be described by a single rheological model, but rather behave as dynamic two-phase Coulomb mixtures (Iverson 1997; Iverson 2003), where flow behaviour is a function of temporal and spatial variations in pore-fluid pressure, mixture agitation (granular temperature), and grain-size distribution. Abrupt vertical, lateral and longitudinal variations in deposit sedimentology and morphology thus reflect the interplay between channel morphology (width and gradient) and the typically unsteady, non-uniform flow behaviour of Ruapehu lahars. In the following discussion lahars and their deposits are designated by the date and a letter representing how the event was initiated; where E = eruption-triggered, L = lake outburst, and R = remobilisation.
283
The cause of the flood was attributed to breaching of an avalanche-dammed lake in the upper gorge, or “to a sudden escape of vapours from the same volcanic source that gives origin to the mineral waters” (Skey 1869), as the existence of the Crater Lake was unknown to European settlers until 1879 AD. No volcanic activity was associated with this event, but an eruption of Ruapehu was witnessed on 16 May 1861 AD although no accompanying lahars were reported (Gregg 1960). The 13 February 1861 AD event is estimated to have been the largest in the historical lahar sequence with a total volume of c. 6 million m3 (Hodgson 1993). An eruption-triggered lahar of that magnitude would have also impacted the Wanganui catchment, but no lahar was reported there. Given the size of the lahar, the lack of an associated eruption, and the absence of lake floor material in the 1861 lahar deposits, it is here inferred that the flow resulted from a break-out of Crater Lake. The 1861 L deposit has a limited exposure 7 km from Crater Lake where scour from the 2007 L lahar exposed approximately 2 m of vertical section in a normally dry channel on the south side of a mid-valley lava outcrop (Fig. 2 and 4). The deposit consists of very poorly sorted, coarse-skewed diamicton with sub-angular boulders, cobbles and pebbles in a sandy silt matrix that comprises 13 wt. % (Table 2). Median clast size is 26.5 cm, and the largest boulders are 1.5 m in diameter. The upper metre of the deposit exhibits crude normal grading associated with a slight increase in matrix content (Fig. 5). Clasts are predominantly brick-red and black scoria and andesitic lithics; fragments of lake sediment, common in eruption-triggered lahar deposits, are conspicuous by their absence. Identification of these deposits as belonging to the 1861 AD event is based on their geomorphic location across a drainage divide that would require a large lahar to cross, and the overlying stratigraphy comprised of ash correlated with the 1945 AD eruption, buried by deposits of the 1975 AD lahar. Within the fan the 1861 L deposits reach 1.2 m thick and are underlain by a 20 cm thick soil horizon (Hodgson et al. 2007). 1945 AD eruption-triggered lahars
1861 AD crater lake break-out lahar The first recorded lahar in the Whangaehu valley was observed near the coast at Whangaehu (150 km from Crater Lake) in February, 1861 AD where it destroyed the recently constructed road bridge (Taylor 1861). “The flood overflowed its banks, depositing vast quantities of snow, and drift along the entire course... in little more than two hours it subsided, leaving large masses of ice, snow, and mud, filled with crystals of ice, on the banks... very black, and emitting a strong sulphureous smell….”
In 1945 AD Mt. Ruapehu experienced its longest historic episode of magmatic activity characterised by the episodic extrusion of a lava dome into the crater basin, and its explosive phreatomagmatic, and later magmatic, disruption (Oliver 1945; Reed 1945; Beck 1950; Johnston et al. 2000). Crater Lake water was progressively displaced by dome growth, with overspill occurring at c. 20 times the normal rate by late June, and was completely emptied by July (Reed 1945). No primary lahars were associated with this eruption, but examination of aerial photographs from 1940 and 1950 show that some channels on the Whangaehu fan were modified as a result of the eruption (Fig. 6a).
0.09
0.08
0.06
0.06
37%b
11%b (variable)
30%
40%
12–50%b (variable)
45% (snow) 55% (silt)
1975E
1995E
1995R
1999R
2007L
2007E
e
d
c
b
a
0.15
13%b
1861L
20
134
108
258
311
450
154
Max. grain size (cm)
2.45
2.80
2.45
2.85
2.78
2.54
2.95
Sortinga
Matrix
Both
Matrix
Both
Both
Matrix
Clast
Support
Horizontal preference of clasts
None observed
None
None
Semi-horizontal large clasts
Semi-horizontal large clasts
No
Imbrication
SR
SR to R
SR
SA
SR
A to SA SR
Clast shapec
Post melting: massive with snow melt structures
Overlapping lenses of open work gravels, gravelly sands, silty gravels Snow rich: two normally graded snow units
Massive with alternating clast and matrix support Massive
Crude normal grading in lower unit Lenses of grain size concentrations Lenses of normal graded sand Massive
Structure
No significant sub-0.016 mm particles were found
HA Hydrothermally altered; A andesite; S scoria; XL = crystal mud component in these samples are silt, particles >0.016 mm in diameter
A angular; SA Subangular; SR Subrounded; R Rounded
Houghton and Fagents (pers. comm)
All values within one standard deviation
0.14
0.07
Mean matrix grain size (cm)
Wt. % matrix (<16mm)
Unit
Table 1 Sedimentology of recognised historic lahar deposits in the upper Whangaehu valley
Hydrothermally-altered clasts Native sulphur Gypsum Pyrite crystals Snow/ice Unconsolidated lake sediment
Hydrothermally-altered clasts Lake sediment, Gypsum, Pyrite crystals Siliceous sinter
Lake sediment and hydrothermally-altered clasts black scoria Altered black scoria
Black scoria
Dominated by red scoria No crater lake sediment Lake sediment
Distinctive lithology
46/24/24/6
25/32/34/10
23/21/52/6
24/21/46/8
17/28/45/10
18/23/60/9
16/18/61/5
Components % HA/A/S/XLd
284 Bull Volcanol (2010) 72:279–296
Bull Volcanol (2010) 72:279–296
285
Fig. 4 Cross-section of the upper Whangaehu valley 7.1 km downstream of Crater Lake at the Round-the-mountain-track. Position of active channel arrowed. a Terrace configuration before the 18 March 2007 break-out lahar (2007L). No 1975E deposits were preserved in the main channel at this point, although they are terraceforming a few 100 m up- and downstream. b Terrace configuration immediately following the 25 September 2007E lahar before melting. Note erosion by 2007L in spillway and deposition in main channel, and associated shift in active channel. Ten times vertical exaggeration
Table 2 Paleohydraulic reconstructions of historic lahars in the upper Whangaehu valley near the Round-the-mountain-track (6.9 km downstream of Crater Lake). Onetapu Formation identifications from Hodgson et al. 2007 Event
Stage (m)
From travel-time data
From local super-elevation data
Interpolateda
n
n
n
Velocity (m/s)
Velocity (m/s)
Discharge (m3/s)
References
1861L One-3
10–17
0.15
9,800–11,000
This study
1953L One-4 1975E One-5 1995E One-6 1995R 1999R 2007L One-7? 2007E
6.1
0.15
1,500–2,000
Manville (2004)
0.11
7,700 5,000 1,400–1,800
This study Nairn et al. (1979) Cronin et al. (2000)
420 220 4,000 1,900–2,300 1,650-1,800 1,000–1,200
This study This study This study Manville and Cronin (2007) This study Manville and Cronin (2007)
a
12 6 5 3.5 2.5 8.3 8.3 4 4
0.078
10 10 9.1
0.11 0.11 0.072
9.67
0.03
13.5
0.11 0.032
12.6–13.5 0.05
Values of channel roughness (n) are interpolated from grain size, morphology, and inundation area of the flow path for a given lahar
286
Bull Volcanol (2010) 72:279–296
(Fig. 4). Mapped inundation of the fan by the 1953 AD lahar is a minimum (Fig. 6b) due to the low preservation potential of highwater marks (mudlines) and the delay between the event and the next aerial photography survey. 1975 AD eruption-triggered lahar
Fig. 5 1861 AD lahar deposit exposed beneath the chronostratigraphic marker of the 1945 AD ashfall and 1975 AD lahar deposit, in cross-section A–B of Fig. 4 (T20/370095)
1953 AD crater lake break-out lahar Following the 1945 AD eruptions, Crater Lake was refilled over the next 8 years from precipitation run-off, snowmelt, and geothermal condensate behind a barrier of unconsolidated tephra, buttressed by 1945 AD lava and the Crater Basin Glacier over the former outlet area. On Christmas Eve 1953 AD the barrier failed, enlarging the existing ice tunnel over the outlet, and releasing c. 1.8 million m3 of water as the lake level dropped by 7.9 m (O’Shea 1954; Manville 2004). The outflow rapidly entrained snow, ice, and volcanic debris, including older lahar deposits, to form a lahar that reached Tangiwai, 39 km downstream, as a debris- to hyperconcentrated flow. The lahar critically damaged the Tangiwai railway bridge minutes before arrival of the Wellington-Auckland express train, resulting in the loss of 151 lives. Paleohydraulic analysis estimates the peak discharge in the upper Whangaehu valley at c. 2,000 m3/s (Manville 2004). Deposits that can be correlated with this event are not recognised in the gorge, but they are mapped farther downstream as member One-4 of the Onetapu Formation (Hodgson et al. 2007). The 1953 AD lahar did not overtop the drainage divide at the RTMT and is not found in the branch south of the mid-gorge lava outcrop
The largest measured historic lahar occurred on April 24, 1975 AD as a result of a phreatomagmatic eruption through Crater Lake. Base surges and ballistic fallout inundated c. 8 km2 of the summit, triggering lahars in multiple catchments including the Whangaehu gorge, where peak discharge was estimated at c. 5,000 m3/s from superelevation and highwater marks (Nairn et al. 1979). Total lahar volume was gauged at 1.8 million m3 57.1 km downstream (Paterson 1980). Lake level fell by 8 m as a result of the eruption, representing a volumetric loss of 23%, but the volume of ejected water expelled cannot be precisely constrained because of snow and ice entrained back into the lake and bathymetric changes associated with the eruption, although it was estimated at up to 50% of the total volume of water returned to the lake (Nairn et al. 1979). Oblique aerial photographs show that the lahar filled the upper Whangaehu valley near the RTMT, with highwater marks visible several metres above the highest depositional surface (Fig. 7). Following incision in subsequent decades, the remnants of the 1975 AD deposits form the highest continuous lahar terrace surfaces in the upper Whangaehu valley, but thin rapidly on the fan (Fig. 6c) to c. 80 cm (Hodgson et al. 2007). The surface of the 1975 AD deposit displays two distinct morphologies: bouldery axial terraces and smooth lateral terraces. Channel axis terraces display large scale boulder bars up to 30 m wide and 170 m long with average outsized clasts 50–310 cm in diameter, and streamlined bedforms averaging 10–15 m in width and 45 m in length. These features are only disturbed where they were undercut by subsequent lahars. In wide (>200 m) stretches of the channel, smoothed lateral terraces up to 25 m wide occur, with smaller outsized clasts consisting of (30–60%) scoria. The 1975 AD deposits are very poorly sorted, comprising matrix- and clast-supported silty pebble-cobble and bouldery gravels with laterally discontinuous lenses of increased coarse concentrations and sorting, reflective of pulses or transformations within the lahar (Fig. 8). Sandsized cubic crystals of pyrite, products of the hydrothermal system in the lake, occur within the matrix. Clasts over 16 mm form <50 wt. % of the deposit, giving it a low median grain-size, but it contains the largest clasts of any historic lahar in the form of boulders of andesite lava up to 4.5 m in diameter that display a subhorizontal upstreamdipping orientation. The petrography of the 1975E deposit is distinctive, with the coarse population dominated by
Bull Volcanol (2010) 72:279–296
287
Fig. 6 Inundation areas of historic lahars along the first 17.4 km of the upper Whangaehu valley and fan reconstructed from archival vertical and oblique aerial photographs. a 1940 channel and 1945E lahar deposits. b 1950 channel and 1953L deposits, minimum inundation due to delay between the flow and the 1954/54
photogrammetric survey and no data within 9 km of Crater Lake. c 1975 channel and 1975E deposits. d 2006 channel and 1995E deposits. e 2006 channel and 2007L deposits. f 2007 channel and 2007E deposits
subequal proportions of red and black andesitic scoria. Well-rounded cobbles of black scoria with uniform, spherical, vesicles ~5 mm in diameter, are diagnostic, and are often concentrated at the outer margins of lateral terrace surfaces, presumably due to hydraulic sorting and rafting owing to their relatively low density. Blocks of pale grey hydrothermally altered lake floor sediments and ventderived cemented breccia also occur, but are fragile and have often crumbled as result of weathering. Wood frag-
ments from the RTMT bridge destroyed by the lahar form useful chronostratigraphic markers downstream, due both to the distinctive size of the timber and the unusual capped nails used in its construction. 1995 AD eruption triggered lahars The 1995 AD eruption of Ruapehu began with a series of escalating phreatic and phreatomagmatic eruptions that
288
Fig. 7 Oblique aerial photograph of fresh 1975AD lahar deposits in the upper Whangaehu valley immediately upstream of the RTMT bridge that was destroyed by the lahar (GNS archive). The deposit surface is covered in large boulder bars and displays evidence of reworking from the recessional limb of the lahar. White arrows indicate high water marks from the 1975E lahar
generated minor snow-rich lahars on 18 and 23 September (Bryan et al. 1996). Semi-continuous explosions on 25 September produced a multi-peaked lahar in the Whangaehu valley, augmented by water displaced over the natural outlet (Cronin et al. 1997b), and the entrainment of large volumes of snow and ice from the summit area. These flows proved to be the largest lahars of the eruptive episode (Fig 9; Table 2), depositing significant bouldery terraces in the upper Whangaehu valley from the passage of non-cohesive
Bull Volcanol (2010) 72:279–296
debris flows with a peak discharge estimated at 1,400– 1,800 m3/s (Cronin et al. 1997b). Although there were 18 eruption-triggered lahars recorded during the 1995 AD sequence, they cannot be distinguished in the field, so it is assumed that preserved deposits represent the largest lahar, LH4 (Cronin et al. 1997b), and are here referred to as the 1995E deposits (Graettinger 2008). The 1995E deposits are constrained by the channel eroded into the larger 1975E terraces by two decades of background flow and the deposits of early eruption-triggered lahars. The 1995E terraces were themselves extensively modified by their own recessional limbs and later rain-triggered lahars, resulting in the rapid reestablishment of the channel following each event. The 1995E terraces have surface morphologies and textures indicative of multiple events and pulses. Deposits are very poorly sorted, mixed matrix- and clast-supported silty gravels with laterally discontinuous lenses of increased coarse-clast concentrations and sorting. Maximum grain size is 3.1 m, but by the apex of the fan the matrix has fined and deposits become sand-dominated, with a decrease in both gravel and mud content further downstream (Cronin et al. 1997b). The deposit has the highest overall silt content of historic units at up to 12 wt. %. Streamlined boulder bedforms are common with frequent surficial silty/muddy pond deposits up to 30 cm thick. Small mounds of highly stratified sand and silt accumulated behind flow obstructions as a result of repeated lahar pulses. Petrographically, the 1995E deposits contain the highest abundance of andesitic lava and fragile hydrothermally altered andesite and sinter of historic lahar units. Black scoria is also common. 1995–1999 AD rain-triggered lahars
Fig. 8 Crude sub-horizontal stratification in deposits of the 1975E lahar reflecting pulses and/or transformation of the lahar. Located c. 9 km from Crater Lake in sample section, East. Scale staff 4 m long (T20/365097)
Crater Lake was emptied by eruptive activity on 12 October 1995 with the eruption moving into a dry, magmatic phase that produced two sub-plinian ash eruptions before activity declined in November. Between October 1995 and April 1996 fresh accumulations of juvenile material were remobilized through precipitation and the collapse of tephraladen snow (Manville et al. 2000). The lahar deposits are treated as the product of a single prolonged depositional event throughout the spring and summer of 1995/96, referred to herein as 1995R. The largest of the remobilisation events occurred on October 28 during a large storm, with rain-triggered lahars occurring in multiple catchments: the flow in the Whangaehu was estimated at half the size of the largest eruption-triggered lahar (Table 2). Remobilisation of the 1995/96 eruption products continued with decreasing frequency and magnitude, with the last recorded event occurring in the austral autumn of 1999 during a rain storm, but there were no observations and limited documentation of this event.
Bull Volcanol (2010) 72:279–296
289
Fig. 9 Repeat photography of a main channel of the upper Whangaehu valley (6 km from Crater Lake) a few days before the 18 March 2007 lake break-out lahar, terraces are composed of 1975E
lahar deposits (T20/363097). b Same area a few days after the 2007L lahar; aggradation totalled 12–14 m. Arrows indicate large boulders undisturbed by the 2007L lahar. (Photos: S. Fagents and R. Carey)
The 1995R deposits are predominantly composed of sand and silt with surfaces dominated by rafted low density hydrothermally altered clasts and scoria, and are inset within the edges of the higher 1995E terraces. Scoria produced on October 12 and 14 had a brown crust that when redistributed and weathered produces a distinctive orange-brown colour on deposit surfaces. The deposits are generally unbedded with minor variation in grain size, sorting and clast/matrix support similar to the those observed in large deposits and related to flow pulses, but instead of a discrete terrace the deposits form a series of steps down to the active channel. Some concentrations of outsized clasts are present on the deposit surface, but there are no discrete bedforms. The 1995R deposits develop minor stratification and grading downstream in the Whangaehu fan 16 km from source (Cronin et al. 1997b). The scoria-rich minor terraces inset below the 1995R surface are identified as products of the 1999R flow. The 1999R deposits are silty sand with abundant (60 wt %) subrounded outsized clasts. Some surficial pond deposits of fines display crude normal grading and lack coarse clasts. Maximum clast size is 110 cm in diameter. The deposit is dominated by low density material, mainly scoria, but also including wood debris from the reconstructed RTMT bridge. The andesitic lava content is the lowest of all the historic lahar deposits. Prior to erosion by the 2007 L lahar the 1999R deposits lined the majority of the active channel with stepped terraces up to 3 m thick, 0.3 to 0.6 m inset below the 1995R surfaces. The 1999R deposits were not identified downstream of the fan.
deposited over the outlet channel by the 1995/96 eruptions (Manville and Cronin 2007). Approximately 1.3 million m3 of water was released as the lake fell by 6 m, eroding and entraining snow, glacial ice, landslide debris, talus material and older lahar deposits along the upper Whangaehu valley to transform to a non-cohesive debris flow by the RTMT. Comparison of pre- and post-lahar LiDAR surveys show that c. 2.4 million m3 of material was stripped from the first 10 km of flowpath, although there was local aggradation of up to 14 m (Fig. 9). Deposit surfaces display a range of bedforms including gravel bars, streamlined cluster bedforms, minor openwork levees and exposed patches of largely buried terraces of 1975E and 1995E deposits. Establishment of a single channel following the lahar was delayed by repeated mobilisation of material along the surface of the deposit. Some bedforms display normal grading and upstream fining, but the majority are loose openwork cluster accumulations. These features are about 60 cm wide and 5 m long. Escaping pore water resulted in the accumulation of silt-sized material above impermeable layers, which and escaped along vertically eroded terrace faces. In the first few months following deposition, dewatering and compaction structures frequently developed. Limited incision of the deposit in 2007 exposed the upper metre, revealing internal sedimentary structures which reflect braided stream-like deposition, indicative of reworking by the dilute tail of the lahar. Multiple blue-grey mud coatings on outsized clasts, erosional scarps and valley walls were plastered up to 3 m above the main deposit surface, and corresponded to the maximum flow stage. Between 0.2 and 1 m above the deposit surfaces this mud coating was largely stripped off by a more dilute portion of the lahar, and a second mud (locally pebble-rich) coating formed only slightly above the
2007 AD crater lake break-out lahar On 18 March 2007, the refilling Crater Lake breached an unstable barrier of unconsolidated ash, lapilli, and blocks
290
Bull Volcanol (2010) 72:279–296
Fig. 10 Frozen deposits of the 2007E lahar the 2 days following emplacement west of the RTMT bridge (same location as pictured in Fig. 7). a The two surface types observed in the 2007E deposit, complex ridges of the first two lahars, and the smoothed surface of the third, dilute lahar that also incised a 20 m wide channel through the entire deposit. b A cross-section of the frozen 2007E deposits: the first
eruption-triggered lahar that travelled over the Whangaehu Glacier (0– 110 cm), the second eruption-triggered lahar that travelled over the normal outlet of Crater Lake, distinguished by a brown colouration from material incorporated from a landslide near the outlet (110– 140 cm), and the dilute late stage lahar caused by displacement over the outlet by adjustment of the hydrothermal system (140–150 cm)
deposit surface (20–30 cm), representing the final concentrated phase of the flow. The mud coatings were mostly eroded within 5 months of the lahar, with those remaining weathered to a dull yellow grey colour. Of the historic lahars, the 2007 L produced deposits with the highest matrix sand content, best rounding, and the largest clast diversity, including scoria, Crater Lake sediments, andesitic lava, and crystals of gypsum, native sulphur and pyrite in the matrix.
modified snow and firn, each coated with a film of silty fines (Fig. 10). Deposit bulk density was 350–660 kg/m3, while the sediment concentration varied between 24–57 wt. % and comprised clay- to boulder-sized lithic fragments. Void space in the frozen deposits was significant at 40–60 vol. %, possibly indicating drainage of a liquid phase from the lahars as they stopped, or immediately after. Individual flow units showed coarse-tail normal grading, with coarse clasts including ice fragments up to 15 cm in diameter and lithic boulders up to 30 cm across. Lithic fragments were dominated by hydrothermally altered andesite lava and scoria, often veined or crusted with gypsum and alunite, and pieces of pale, laminated lake floor sediment. Black mmsized globules of native sulphur were abundant in the matrix. The 2007E deposits had the highest hydrothermal-clast content of the historic lahar sequence. The surface texture of the early lahar deposits was coarse, rugose, with curved pressure ridges developed on a 10–20 cm wavelength, marginal flow-parallel shear structures, and steep lobate margins where they onlapped older terrace surfaces. The third lahar modified and partially eroded the early lahar deposits, producing a series of initially thin and superficial lithic-poor and coarse fragment-depleted ice slurry deposits with a locally distinctive smooth and pahoehoe-like surface texture, that became more obvious downstream of the RTMT. These deposits terminated c. 15 km from source, but with intermittent traces of rafted snow deposits continuing as far as the Wahianoa aqueduct (23.8 km). The third lahar and background streamflow had incised a c. 20 m wide channel through all the ice slurry deposits by the morning of 26 September. Within a month, thawing of the frozen deposits had led to c. 90% reduction in thickness and
2007 AD eruption-triggered lahars On 25 September 2007, a small phreatic eruption through Crater Lake generated a number of primary ice slurry lahars (cf. Pierson and Janda 1994; Waitt et al. 1994; Cronin et al. 1996) in the upper Whangaehu valley (Lube et al. 2009). Two closely-spaced flow pulses resulting from run-off of ejected Crater Lake water, ballistics, and entrained snow and ice from the summit plateau, ran down the Whangaehu glacier before entering the gorge, while a second, slightly smaller lahar was generated by syn-eruptive water displacement across the lake outlet, coupled with a topographically controlled base surge into the outlet channel. A small ice slurry lahar also occurred on the north side of the mountain in the Whakapapaiti catchment, entering the Whakapapa Skifield. A third Whangaehu valley lahar apparently was triggered by passive volumetric displacement of water from Crater Lake c. 1.5 h after the eruption as a result of re-adjustment of the sub-lake hydrothermal system. At the RTMT, the eruption-triggered lahars deposited 1–3 m of ice slurry with a matrix composed of millimetresized sub-rounded particles of clear ice, representing
Bull Volcanol (2010) 72:279–296
preservation as a thin veneer (0.15–0.7 m) of highly vesicular diamict.
Lahar magnitude The magnitude of lahars is typically expressed in terms of their peak discharge and total volume. Matters are complicated by the bulking and debulking of lahars as they entrain and then deposit particulate material, so that where the flow is measured is almost as important as what is measured. Lahars are seldom at their largest at conventional or convenient gauging sites, but it is here inferred that most Ruapehu flows reach their maximum sediment concentration and discharge and volume just after the steepest portion of the upper Whangaehu valley, in the vicinity of the RTMT, so that downstream complications from debulking, and effects of attenuation during flow conveyance. Peak discharge is the product of the cross-sectional flow area, derived from channel geometry and high stage marks, and the local mean flow velocity, which is obtained by a variety of methods, including: direct observations, particle-image velocimetry, super-elevation marks (Johnson and Rodine 1984), arrival times, or numerical simulations. Where only channel geometry and stage heights are known, velocity may be estimated using Manning’s equation and an interpolated value of channel roughness (n) (Chow 1959; Table 2). In general, peak discharge scales with volume (Vignaux and Weir 1990; Pierson 1998), but the shape of the starting hydrograph is also important. Eruption-triggered lahars such as the 1975E event are generated more or less instantaneously, resulting in a very sharply peaked, impulsive hydrograph, whereas lake break-out and displacement lahars show a broader hydrograph with a lower peak. Impulsive lahars tend to attenuate more rapidly due to increased frictional interactions with the channel so that although larger proximally, they are smaller downstream than other types of flow. The shape of the hydrograph also is indicative of bulking potential and deposit sedimentology, with lahars having rapidly rising hydrographs producing thicker and more massive proximal deposits (Rushmer 2007). Influences on lahar magnitude Historic lahars at Ruapehu span several orders of magnitude (Table 2), and extending the record back to c. 2 ka adds another two orders of magnitude (Hodgson et al. 2007). This range reflects both the diversity of lahartriggering processes and external environmental factors. Lahar triggering processes at Ruapehu include eruptions that eject Crater Lake water onto the upper slopes of the
291
mountain (Nairn et al. 1979), lake displacement events due to progressive uplift of the crater floor (Vignaux and Weir 1990), lake break-out events following temporary impoundment of excess water behind ephemeral barriers (O’Shea 1954; Manville 2004; Manville and Cronin 2007), and raintriggered lahars where tephra deposits are destabilised by heavy rain (Hodgson and Manville 1999). The trigger mechanism primarily controls the water budget for lahar formation, both in terms of volume and supply rate, whereas the bulking of the flow is more reflective of the sediment budget. Eruption-triggered lahars For the present configuration of Ruapehu’s active vent under Crater Lake, the starting volume of an eruptiontriggered lahar depends on the volume of water in the Crater Lake, which has varied historically between c. 7 and 11 million m3 as volcanic activity has modified both the shape and depth (70–180 m) of the lake basin, and the depth and magnitude of the volcanic explosion (Wohletz 1986; Koyaguchi and Woods 1996). Multiple closelyspaced explosions can generate compound lahar waves, as with LH4a-e of the 1995 lahar sequence (Cronin et al. 1997b). Nairn et al. (1979) suggested that the 24 April 1975 eruption instantaneously ejected c. 50% of Crater Lake, using the shallow marine nuclear bomb test at Bikini Atoll (Moore 1967) as an analogue for Crater Lake’s base surges. About half of this water, however, fell or drained back into the lake basin. Additionally, this drain-back would have entrained an unknown and potentially very substantial volume of snow and ice from the Crater Basin catchment, and changes in lake bathymetry as a result of the explosion are also unknown. Because of these complications, lowered post-eruption lake levels are poor indicators of the size of the eruption and volume of water expelled. Otway et al. (1995) considered that the largest credible eruption at Ruapehu based on the post-2 ka sequence would eject a maximum 50% of the lake outside the Crater basin. Partial or complete collapse of the hard-rock rim of Crater Lake, in conjunction with, or independent of an eruption, has therefore been invoked to explain the large magnitude of some of the post-2 ka Onetapu Formation lahars, which is supported by high proportions of hydrothermally altered material (mainly clays) in the deposits (Lecointre et al. 2004; Hodgson et al. 2007). The topography of the Crater basin is such that prehistoric configurations of Crater Lake could have held much more water than today. Moreover, simple numerical models show that as explosions become more powerful, not only more water is expelled but, a greater proportion of it lands outside the Crater basin, so that complete ejection of Ruapehu’s Crater Lake is a possibility (Mastin and Witter 2000). Such large scale
292
Bull Volcanol (2010) 72:279–296
crater lake emptying has occurred at other volcanoes, such Mount Kelud, in Indonesia where the entire 30–40 million m3 crater lake was emptied in the 1919 AD eruption (Scrivenor 1929). Additionally, explosions at greater water depths (up to a critical depth, estimated at a several 100’s of metres (Mastin and Witter 2000), where the explosions become suppressed) transfer energy to the water column more efficiently, resulting in wider dispersal of displaced water (Wohletz 1986; Mastin and Witter 2000). Uplift of the crater floor by rising magma or lava dome growth can also displace water over the natural outlet of the crater lake causing redistribution of lahar terraces (1945), or discrete lahars (1968), but the rate is likely to be too low to generate concentrated lahars. The volume of particulate ejecta varies greatly with the eruptions’ size and type. Magmatic and phreatomagmatic eruptions are more likely to expel a larger volume that includes juvenile material and vent-wall accidental lithics; these are in addition to the disrupted lake floor sediments and hydrothermal debris that are associated with purely phreatic eruptions. Juvenile material forms a significant fraction of the 1975E deposits, for example.
1999). A number of similar or larger flows were triggered in the Whangaehu catchment between 14 October and 1 December 1995. Deposition of tephra layers on a sloping ice or snow substrate, or the presence of internal discontinuities such as low permeability fine-ash layers, enhance the remobilisation potential. Destabilisation is caused by infiltration of rainfall, which increases pore fluid pressures, reduces intergranular friction, and increases deposit density until shear failure and subsequent liquefaction occurs (Hodgson and Manville 1999). Rain-triggered lahars on Ruapehu begin as a liquefied mass, often of low density scoria, and are typically small. The maximum size of raintriggered lahars appears to be proportional to the time interval between rainfall events of sufficient intensity to trigger failure (Caine 1980; Wieczorek 1987). Frequent rainfall decreases the potential for tephra preservation on steep slopes, although rain-generated lahars have been recorded up to 6 years post-eruption at Ruapehu (the pre1995 AD record is sparse). Because of the predominance of low density material in remobilisation lahars, their bulking potential through erosion also appears limited.
Lake break-out lahars
A major factor governing the magnitude of a lahar is its capacity for volumetric bulking, which reflects both the erosive power and duration of the flow and the availability of material for entrainment. Bulking can be discussed in terms of apparent bulking factor (ABF), including only the volume of added solids (Vallance and Scott 1997), or total volumetric bulking, referring to the addition of any volume to the lahar (water, ice, sediment). Erosive power includes both flow-capacity and flow-competence factors (Hiscott 1994; Manville and White 2003), and may be expressed in terms of maximum boundary shear stress or stream power (W/m2; Baker and Costa 1987). The multi-metre diameter boulders entrained and transported in the upper Whangaehu valley by historic lahars indicate that competence is rarely a limiting factor. If a lahar entrains water in the form of snow and ice, over-run ambient stream water, or interstitial porewater from the eroded substrate then the volumetric bulking factor can increase, as the capacity increases, with values of 15–20 reported for some lahars (Gallino and Pierson 1985). Lateral entrainment through undercutting of talus slopes and older alluvial and laharic terraces along the channel margin is significant in many lahars. Bed entrainment can also occur through the transmission of basal shear stress to the bed, turbulence and lift, localised grain-bed interactions (i.e. impact and scour), and generation of excess pore-water pressures through rapid loading (Gauer and Issler 2004; McDougall and Hungr 2004). Based on comparison of preand post-2007 L lahar DEM models derived from LiDAR data, the in-channel volume deficit (excluding lateral
The initial volume of lake break-out lahars is a function of the diameter of Crater Lake and the elevation difference between the maximum surface elevation of the lake and the base of the breach that develops. The 1953 AD break-out released an estimated 1.8 million m3 of water as lake level fell by 7.9 m, while the 18 March 2007 breakout released an estimated 1.37 million m3 with a head loss of 6.2 m. The surface area of Crater Lake increases as the square of water depth above the level of the hard lava sill at the outlet (Christenson and Wood 1993). The initial peak discharge is determined by the rate of breach formation which is dependent on the material and geometric properties of the dam, the excess head in the lake, and the shape of the breach. Peak outflow rates of c. 400 and 540 m3/s have been calculated for the 1953 L and 2007 L break-outs (Manville 2004; Manville and Cronin 2007). Rain-triggered lahars The initial volume of rain-triggered lahars is a function of the thickness and area of tephra remobilised, and the volume of entrained water in the particle interstices. One such event on 28 October 1995 remobilised a tephra layer between 20–80 cm thick from a 0.7 km2 area of the Mangatoetonui glacier, generating a lahar with a minimum estimated volume of 330,000 m3 and a peak discharge of c. 1,400 m3/s (Manville et al. 1998; Hodgson and Manville
Controls on bulking
Bull Volcanol (2010) 72:279–296
contributions from syn-lahar landslides) in the upper Whangaehu valley between Crater Lake and the RTMT is c. 1.2 million m3, corresponding to a channel debris yield rate of c. 175 m3/m, or an average of 3 m of vertical scour (Manville et al. in preparation). Since most of the gorge is inferred to be floored by solid lava beneath a veneer of alluvium and lahar deposits, the gorge effectively limits the sediment supply to passing lahars. For this reason, the time interval between lahar events and/or eruptions may be significant in determining the volume of material available for entrainment (Jakob et al. 2005), through the need to recharge the gorge floor with talus, landslide, and other volcanic material. Landslides triggered by passing lahars may be a significant contributor to the bulking of some lahars: a number of small failures can be observed on archival aerial photographs along the margins of the 1975E lahar path while the March 2007 break-out lahar (re-)activated a very large landslide (volume deficit c. 900,000 m3) in the upper Whangaehu gorge, in an area that had experienced large scale failures in the past. The amount of snow and ice available for entrainment, primarily by eruption-triggered lahars, depends on the time of year and the dryness of the winter; average snow depths at Ruapehu reach c. 2.5 m at elevations above 2,000 m during September and are minimal, to zero, during the late Austral summer. The September 1995 and 2007 eruptiontriggered lahars transformed into ice slurries, dominated by entrained winter snow, in contrast to the April 1975E lahar deposits which were almost snow-free. Chemical data show that ice slurry lahars can entrain 10–100 times their starting crater lake water volume in snow and firn (Lube et al. 2009). Snow and ice in the constricted upper Whangaehu valley is a less significant source for lahars exiting through the lake outlet, in particular as the Whangaehu and Crater Basin glaciers have receded over the past 50 years (Krenek 1959; Keys 1988).
Discussion The sedimentology, petrography and geomorphology of mass flow deposits in the upper Whangaehu valley provide insight into the triggering mechanisms, flow behaviour, bulking history and depositional processes of historic lahars. Trigger mechanisms have been established for the majority of these, informing analysis of their sedimentary, petrologic and geomorphic characteristics. The lessons learnt here can be applied to the interpretation of prehistoric deposits in order to improve estimates of event magnitudefrequency. For events from which no deposits are preserved, such as the 1925 event, the trigger mechanism cannot be reconstructed.
293
Deposit componentry Sediment supply in the upper Whangaehu valley is limited to juvenile debris, hydrothermally altered material, diagenetic minerals, and lake floor sediments ejected from the Crater Lake, snow and ice removed from the Whangaehu Glacier and valley, and material eroded from the valley walls and floor, including lava flows, talus slopes, landslides, and older lahar deposits. Five principal components are have contributed to historic lahar deposits: andesite lava, scoria, sand- to silt-sized diagenetic minerals, hydrothermally altered clasts and snow/ice (Table 2); clay-matrix dominated deposits from sector collapses of hydrothermally altered parts of the volcanic edifice are restricted to the prehistoric record: (Hodgson et al. 2007). Recycling of older lahar deposits by young, large lahars complicates the reconstruction of trigger mechanisms from deposit componentry, but relative proportions offer some guidance. Abundant fragments of lake floor sediment and hydrothermally altered and cemented material, including sulphur globules and gypsum in the matrix are indicative of eruption-triggered lahars, where the vent system and overlying bed of Crater Lake have been explosively disrupted. Up to 2 mm pyrite crystals are also found in the matrix of the 2007 L deposits, suggesting a post-lahar diagenetic origin, such as precipitation from escaping crater lake water. The absence or under-representation of these components indicates a non-eruptive lake break-out event, as is inferred for the 1861 AD lahar. Deposits of historic rain-triggered lahars are finer-grained and dominated by unaltered low density scoria and lake bed fragments, reflecting a lahar-source limitation in clast size and type, and lack pyrite and gypsum. Coarse, bouldery deposits are correlated with large magnitude, high energy lahars, either eruption-triggered or lake break-out, and only very large historic lahars have overtopped a drainage divide at the RTMT to enter a southern channel (Fig. 4). The primary route of large eruption-triggered lahars off the Whangaehu Glacier directs them across a large, relatively young, icecontact lava flow c. 1.5 km from Crater Lake and into the normally dry South Branch of the upper Whangaehu valley (a route also taken by large rain-triggered lahars sourced on the Whangaehu Glacier in 1995 AD (Fig. 3; Manville et al. 2000)). Large eruption-triggered lahars are thus sub-equally apportioned between the North and South branches of the upper Whangaehu valley, whilst lahars issuing from the lake outlet are largely confined to the North Branch with only a small proportion of flow escaping over the drainage divide. Lahars that have spilled over into the South Branch contain an abundance of black, glassy, blocky andesite clasts showing prismatic fracture patterns, which are inferred to have been stripped from the ice contact lava in the flowpath. While poorly represented in most historic
294
lahar deposits, these form a distinctive component of the c. 450 yr BP Ond-3 member of the Onetapu Formation mapped on the ring-plain (Lecointre et al. 2004; Hodgson et al. 2007), where they were identified as juvenile volcanic bombs. Scattered clasts of this ice-contact lava perched on lava outcrops in the RTMT area c. 15 m above the channel floor may indicate the minimum peak stage level of the 450 yr BP lahar (Fig. 4). Preservation potential The preservation potential of lahar deposits in the upper Whangaehu valley is low, due to the rate of background reworking by stream flow and the rapid erosion of the steep unvegetated environment, and the overprinting effects, and erosion, from large lahars that recur on a c. 20–50 year timescale and reset the terraces in the valley. Only 9 of the c. 46 events known to have occurred in the past 150 years can be distinguished in the RTMT area, and of these, 2 occurred during the course of this study. Preservation is skewed towards the largest and most widely distributed lahars, which form terraces well above the active channel or in side valleys separated from it by drainage divides. Historic lahar deposits have typically filled the active channel (e.g. 1975E; Fig. 7) and then through recessional limb erosion, and further lahar activity the deposits were reduced to terraces that broaden in wider portions of the valley and behind obstructions and constrictions in the flow path (Graettinger 2008). The largest events also tend to produce the coarsest deposits which are the most resistant to erosion. Although the second largest in the historical sequence, the 2007 L lahar did not overtop the 1975E terraces along much of the upper Whangaehu valley: where it did, it was too shallow to rework the boulder bar tops. Deposits of smaller, finer-grained, and active-channelconfined lahars have very low preservation potential: 1995R and 1999R deposits were completely overprinted by the 2007 L lahar in the North Branch of the upper Whangaehu valley, but survived in the South Branch because only a small proportion of the 2007 L lahar entered that catchment. Even the 1953 L break-out lahar deposits cannot be detected in the RTMT area, despite it being the 4th largest in the historical sequence (Table 2), because of overprinting by the 1975E event. Some lahars have significant ice and snow content, such as the LH1-3 events of 1995 AD (Cronin et al. 1996) and the 2007E (Lube et al. 2009). These flows have almost zero preservation potential. Although the flows and initial deposits may be quite thick, melting of the ice particles leaves behind a relatively thin porous deposit that often lacks coarse clasts because of the low density, damped viscoplastic rheology, and low erosive and transport potential of the parent flow (Lube et al. 2009). Similar
Bull Volcanol (2010) 72:279–296
eruption-triggered lahars are hence inferred to be grossly underrepresented in the geological record. As the volume of snow and ice on Ruapehu was significantly greater in the past, it is reasonable to infer that snow slurry lahars and snow collapse-related mobilisation events were more frequent. Due to the inconsistent and limited nature of observations prior to the 1940’s, the frequency of these events is unknown, especially of those too small to reach the ring-plain. Cronin et al. (1997b) only recorded lahars that were observed by chance on the volcanic flank, or which reached the Karioi stream gauge: numerous smaller events were likely confined to the upper Whangaehu valley, as were the 1996E and 2007E flows. Recent upgrades to lahar monitoring at Ruapehu in response to the threat posed by the 2007 L event means that fewer of these events will be missed in the future. Rain-triggered lahars are also typically small and their deposits have a low preservation potential. The size of Whangaehu catchment lahars has been smaller since 450 yr BP, with the deposits of younger flows confined to incised channels on the ring-plain and further downstream (Hodgson 1993; Lecointre et al. 2004; Hodgson et al. 2007). Prehistoric lahars mapped further downstream in the post-1.8 ka Onetapu Formation extend this range by another 1–2 orders of magnitude (Hodgson et al. 2007), yet their deposits, with the potential exception of the 450 yr BP flow, are not recognised in the gorge. Thus, both the relatively infrequent very large, and very common but small, lahars are under-represented in the gorge depositional record. Events between the 1995 and 2007 AD eruptions show that products of both relatively smallscale rain-triggered tephra-remobilisation lahars, and even of large primary, eruption-triggered ice slurry flows, have virtually zero preservation potential. Medial and distal areas will need to be investigated in greater detail to determine if deposits from any of these small events can be identified, based on their stratigraphic relations and internal textures, but the short run out distance of such flows makes preservation downstream unlikely.
Conclusions Over the past 150 years lahars at Ruapehu have spanned two orders of magnitude in terms of peak discharge and volume, both of which have been inferred to reach maxima in the upper Whangaehu valley due to entrainment and bulking by erosion of older lahar terraces, talus fans, landslide deposits, glacial ice and seasonal snow. The diversity of lahar triggering processes is reflected in the componentry of their deposits, whose sedimentological complexity is indicative of the unsteady non-uniform behaviour of the parent flow and its interaction with the
Bull Volcanol (2010) 72:279–296
irregular valley floor and walls. While there have been 46 historic lahars documented within the Whangaehu Valley, only 9 of the deposits have been recognised in the proximal (gorge) and medial (fan) environment. A comparison of the historic record with known pre-historic deposits highlights an underestimate of the potential magnitude of lahars in the Whangaehu valley based on observed lahars alone. Prehistoric tephra and lahar deposits suggest greater sediment budgets, through larger juvenile production or sector collapses. Recorded seasonal and climactic variations suggest greater water budgets, relating to larger snow and lake volumes during an eruptive episode. These findings suggest the necessity of re-evaluating the variability, frequency, and magnitude of lahars in the Whangaehu valley and other catchments at Ruapehu. Acknowledgements This paper is the result of MSc research conducted by A. Graettinger at the University of Waikato. S. Cooke, K. Hodgson, K. Jackson, K. Kataoka, J. Krippner, R. Pickett and M. Taylor provided assistance with fieldwork. B. Houghton and S. Fagents (University of Hawai’i) provided field-sieve data. LiDAR data was used with the permission of GNS Science and Massey University. Manville acknowledges funding from the New Zealand Foundation for Research Science and Technology (CO5X0006). T.C. Pierson and J.D.L. White are thanked for their constructive reviews that improved the quality of this manuscript.
References Baker VR, Costa JE (1987) Flood power. In: Mayer L, Nash D (eds) Catastrophic flooding. Allen & Unwin, Boston, pp 1–21 Beck AC (1950) Volcanic activity at Mt. Ruapehu from August to December, 1945. NZ J Sci Tech B31:1–13 Brayshaw AC (1984) Characteristics and origin of cluster bedforms in coarse-grained alluvial channels. In: Koster EH, Steel RJ (eds) Sedimentology of gravels and conglomerates. Can Soc Petrol Geol Memoir 10:75–85 Bryan C, Christenson BW, Hodgson KA, Houghton BF, Johnston DM, Jurado-Chichay Z, Lindsay J, Manville V, Nakagawa M, Nairn IA, Otway PM, Sherburn S, Scott BJ, Thordarson T, Wood CP, Wilson CJN (1996) Volcanic eruption at a New Zealand ski resort prompts re-evaluation of hazards. EOS, Trans AGU 77:189–191 Caine N (1980) The rainfall intensity—duration control of shallow landslides and debris flows. Geograf Annal 62A:23–27 Campbell IB (1973) Recent aggradation in Whangaehu valley, central North Island, New Zealand. NZ J Geol Geophys 16:643–649 Carling PA (1989) Hydrodynamic models of boulder berm deposits. Geomorph 2:319–340 Carrivick JL, Manville V, Cronin SJ (2008) A fluid dynamics approach to modelling the 18th March 2007 lahar at Mt. Ruapehu, New Zealand. Bull Volcanol 71:153–169. doi:10.1007/s00445-0080213-2 Chow VT (1959) Open channel hydraulics. McGraw-Hill, New York Christenson BW, Wood CP (1993) Evolution of a vent-hosted hydrothermal system beneath Ruapehu Crater Lake, New Zealand. Bull Volcanol 55:547–565 Costa JE (1988) Rheologic, geomorphic, and sedimentologic differentiation of water floods, hyperconcentrated flows and debris
295 flows. In: Baker VR, Kochel RC, Patton PC (eds) Flood geomorphology. Wiley, New York, pp 113–122 Cronin SJ, Neall VE, Lecointre JA, Palmer AS (1996) Unusual “snow slurry” lahars from Ruapehu volcano, New Zealand, September 1995. Geology 24:1107–1110 Cronin SJ, Hodgson KA, Neall VE, Palmer AS, Lecointre JA (1997a) 1995 Ruapehu lahars in relation to the late Holocene lahars of Whangaehu River, New Zealand. NZ J Geol Geophys 40:507–520 Cronin SJ, Neall VE, Lecointre JA, Palmer AS (1997b) Changes in Whangaehu River lahar characteristics during the 1995 eruption sequence, Ruapehu volcano, New Zealand. J Volcanol Geotherm Res 76:47–61 Cronin SJ, Neall VE, Palmer AS (1997c) Lahar history and hazard of the Tongariro River, Northeastern Tongariro Volcanic Centre, New Zealand. NZ J Geol Geophys 40:383–394 Cronin SJ, Neall VE, Lecointre JA, Palmer AS (1999) Dynamic interactions between lahars and stream flow: a case study from Ruapehu volcano, New Zealand. Geol Soc Am Bull 111:28–38 Cronin SJ, Lecointre JA, Palmer AS, Neall VE (2000) Transformation, internal stratification, and depositional processes within a channelised, multi-peaked lahar flow. NZ J Geol Geophys 43:117–128 Donoghue SL, Neall VE (2001) Late Quaternary constructional history of the southeastern Ruapehu ring plain, New Zealand. NZ J Geol Geophys 44:439–466 Donoghue SL, Neall VE, Palmer AS (1988) Holocene geology of the upper Whangaehu River, Mt. Ruapehu. NZ Geol Soc Misc Publ 41A:61 Donoghue SL, Neall VE, Palmer AS, Stewart RB (1997) The volcanic history of Ruapehu during the past 2 millenia based on the record of Tufa Trig tephra. Bull Volcanol 59:136–146 French JR (2003) Airborne LiDAR in support of geomorphological and hydraulic modelling. Earth Surf Proc Landf 28:321–335 Gallino GL, Pierson TC (1985) The 1980 Polallie Creek debris flow and subsequent dam-break flood, East Fork Hood River basin, Oregon. US Geol Surv Water-Supply Pap 2273:21–22 Gauer P, Issler D (2004) Possible erosion mechanisms in snow avalanches. Ann Glaciol 38:384–392 Graettinger AH (2008) Depositional record of historic lahars in the Whangaehu Gorge, Mt. Ruapehu. MSc thesis. Waikato University, Hamilton Gregg DR (1960) The geology of Tongariro subdivision. NZ Geol Surv Bull 40:1–152 Hackett WR, Houghton BF (1989) A facies model for a Quaternary andesitic composite volcano: Ruapehu, New Zealand. Bull Volcanol 51:51–68 Healy J, Lloyd EF, Rushworth DEH, Wood CP, Glover RB, Dibble RR (1978) The eruption of Ruapehu, New Zealand on 22 June 1969. NZ Dep Sci Ind Res Bull 224:1–80 Hiscott RN (1994) Loss of capacity, not competence, as the fundamental process governing deposition from turbidity currents. J Sed Res A64:209–214 Hodgson KA (1993) Late Quaternary lahars from Mount Ruapehu in the Whangaehu River valley, North Island, New Zealand. PhD thesis, Massey University, Palmerston North Hodgson KA, Manville V (1999) Sedimentology and flow behaviour of a rain-triggered lahar, Mangatoetoenui Stream, Ruapehu volcano, New Zealand. Geol Soc Am Bull 111:743–754 Hodgson KA, Lecointre J, Neall VE (2007) Onetapu Formation: the last 2000 yr of laharic activity at Ruapehu volcano, New Zealand. NZ J Geol Geophys 50:81–99 Houghton BF, Latter JH, Hacket WR (1987) Volcanic hazard assessment for Ruapehu composite volcano, Taupo Volcanic Zone, New Zealand. Bull Volcanol 49:737–751 Iverson RM (1997) The physics of debris flows. Rev Geophys 35:245–296 Iverson RM (2003) The debris-flow rheology myth. In: Rickenmann D, Chen C-L (eds) Debris-flow hazards mitigation: mechanics,
296 prediction, and assessment. Millpress, Rotterdam, Davos, Switzerland, pp 303–314 Jakob M, Bovis M, Oden M (2005) The significance of channel recharge rates for estimating debris-flow magnitude and frequency. Earth Surf Proc Landf 30:755–766 Johnson AM, Rodine JR (1984) Debris flow. In: Brunsden D, Prior DB (eds) Slope instability. Wiley, New York, pp 257–361 Johnston DM, Houghton BF, Neall VE, Ronan KR, Paton D (2000) Impacts of the 1945 and 1995–1996 Ruapehu eruptions, New Zealand: an example of increasing societal vulnerability. Geol Soc Am Bull 112:720–726 Keys H (1988) Survey of the glaciers on Mt. Ruapehu, Tongariro National Park - a baseline for detecting effects of climate change. Dept of Conservation, Wellington. Sci Res Int Rep 24:1–20 Koyaguchi T, Woods AW (1996) On the formation of eruption columns following explosive mixing of magma and surface water. J Geophy Res 101:5561–5574 Krenek LO (1959) Changes in the glaciers of Mt Ruapehu in 1955. NZ J Geol Geophys 2:643–653 Lecointre J, Hodgson KA, Neall VE, Cronin SJ (2004) Lahartriggering mechanisms and hazard at Ruapehu volcano, New Zealand. Nat Haz 31:85–109 Lube G, Cronin SJ, Procter JN (2009) Explaining the extreme mobility of volcanic ice-slurry flows, Ruapehu volcano, New Zealand. Geology 37:15–18 Major JJ, Newhall CG (1989) Snow and ice perturbation during historical volcanic eruptions and the formation of lahars and floods. Bull Volcanol 52:1–27 Manville V (2004) Palaeohydraulic analysis of the 1953 Tangiwai lahar: New Zealand’s worst volcanic disaster. Acta Vulcanol 16 (1/2):137–152. doi:10.1400/19101 Manville V, White JDL (2003) Incipient granular mass flows at the base of sediment-laden floods, and the roles of flow competence and flow capacity in the deposition of stratified bouldery sands. Sed Geol 155:157–173 Manville V, Cronin SJ (2007) Break-out lahar from New Zealand’s Crater Lake. EOS Trans AGU 88:441–442 Manville V, Hodgson KA, White JDL (1998) Rheological properties of a remobilised-tephra lahar associated with the 1995 eruption of Ruapehu Volcano, New Zealand. NZ J Geol Geophys 41:157– 164 Manville V, Hodgson KA, Houghton BF, Keys JRH, White JDL (2000) Tephra, snow and water: complex sedimentary responses at an active, snow-capped stratovolcano, Ruapehu, New Zealand. Bull Volcanol 62:278–293 Manville V, Hodgson KA, Nairn IA (2007) A review of break-out floods from volcanogenic lakes in New Zealand. NZ J Geol Geophys 52:131–150 Marutani T, Yamada T, Kimura M, Maita H, Manville V, Leonard G, Trustrum N (2007) Ruapehu Crater Lake break-out lahar: North Island, New Zealand. Shin-Sabo 60:59–65 Mastin LG, Witter JB (2000) The hazards of eruptions through lakes and seawater. J Volcanol Geotherm Res 97:195–214 McArthur JL, Shepherd MJ (1990) Late Quaternary glaciation of Mt Ruapehu, North Island, New Zealand. J Roy Soc NZ 20:287–296 McDougall S, Hungr O (2004) Dynamic modelling of entrainment in rapid landslides. Can Geotech J 41:1437–1448 Moore JG (1967) Base surge in recent volcanic eruptions. Bull Volcanol 30:337–363 Nairn IA, Wood CP, Hewson CAY (1979) Phreatic eruptions of Ruapehu: April 1975. NZ J Geol Geophys 22:155–173 Nakagawa M, Wada K, Thordarson T, Wood CP, Gamble JA (1999) Petrological investigations of the 1995 and 1996 eruptions of Ruapehu volcano, New Zealand: formation of discrete and small magma pockets and their intermittent discharge. Bull Volcanol 61:15–31
Bull Volcanol (2010) 72:279–296 O’Shea BE (1954) Ruapehu and the Tangiwai disaster. NZ J Sci Tech B36:174–189 Oliver RL (1945) Further activity of Mount Ruapehu, May–July, 1945. NZ J Sci Tech B26:24–33 Otway PM, Hodgson KA, Nairn IA (1995) Whakapapa skifields lahar study. Institute of Geological and Nuclear Sciences. Sci Rep 95 (39):1–90 Palmer BA (1991) Holocene lahar deposits in the Whakapapa catchment, northwestern ring plain, Ruapehu volcano (North Island, New Zealand). NZ J Geol Geophys 34:177–190 Palmer BA, Purves AM, Donoghue SL (1993) Controls on accumulation of a volcaniclastic fan, Ruapehu composite volcano, New Zealand. Bull Volcanol 55:176–189 Paterson BR (1980) The hazard of lahars to the Tongariro Power Development, New Zealand. Proceedings, Third Australian-NZ Geomech Conf, NZ Inst Eng, 6(1)/G:2.7–2.14 Pierson TC (1998) An empirical method for estimating travel times for wet volcanic mass flows. Bull Volcanol 60:98–109 Pierson TC, Scott KM (1985) Downstream dilution of a lahar: transition from debris to hyperconcentrated streamflow. Water Res Res 21:1511–1524 Pierson TC, Costa JE (1987) A rheologic classification of subaerial sediment-water flows. In: Costa JE, Wieczorek GF (eds) Debris flows/avalanches: process, recognition, and mitigation. Geol Soc Am Rev Eng Geol 7:1–12 Pierson TC, Janda RJ (1994) Volcanic mixed avalanches: a distinct eruption triggered mass-flow process at snow-clad volcanoes. Geol Soc Am Bull 106:1351–1358 Reed JJ (1945) Activity at Ruapehu, March–April, 1945. NZ J Sci Tech B26:17–23 Rushmer EL (2007) Physical-scale modelling of jökulhlaups (glacial outburst floods) with contrasting hydrograph shapes. Earth Surf Proc Landf 32:954–963 Scott KM (1988) Origins, behavior, and sedimentology of lahars and lahar-runout flows in the Toutle-Cowlitz River system. US Geol Surv Prof Paper 1447-A:1–74 Scrivenor JB (1929) The mudstreams (lahars) of Gunong Kelot in Java. Geol Mag 66:433–434 Skey W (1869) On the water from the Whangaehu River, Onetapu, Auckland. Trans NZ Inst 1:28 Tanguy J-C, Ribière C, Scarth A, Tjetjep WS (1998) Victims from volcanic eruptions: a revised database. Bull Volcanol 60:137–144 Taylor R (1861) Journal of Rev. Richard Taylor 1833–1873. Typescript in Alexander-Turnbull Library, Wellington Vallance JW, Scott KM (1997) The Osceola Mudflow from Mount Rainier: sedimentology and hazard implications of a huge clayrich debris flow. Geol Soc Am Bull 109:143–163 Vignaux M, Weir GJ (1990) A general model for Mt. Ruapehu lahars. Bull Volcanol 52:381–390 Villamor P, Van Dissen R, Alloway BV, Palmer AS, Litchfield N (2007) The Rangipo fault, Taupo rift, New Zealand: An example of temporal slip-rate and single-event displacement variability in a volcanic environment. Geol Soc Am Bull 119:529–547 Waitt RB, Gardner CA, Pierson TC, Major JJ, Neal CA (1994) Unusual ice diamicts emplaced during the December 15, 1989 eruption at Redoubt Volcano, Alaska. J Volcanol Geotherm Res 62:409–428 Wieczorek GF (1987) Effect of rainfall intensity and duration on debris flows in central Santa Cruz Mountains, California. In: Costa JE, Wieczorek GF (eds) Debris flows/avalanches: process, recognition, and mitigation. Geol Soc Am Rev Eng Geol 7:93–104 Witham CS (2005) Volcanic disasters and incidents: a new database. J Volcanol Geotherm Res 148:191–233 Wohletz KH (1986) Explosive magma-water interactions: thermodynamics, explosion mechanisms, and field studies. Bull Volcanol 48:245–264