Bull Volcanol (1998) 59 : 233244

Q Springer-Verlag 1998

ORIGINAL PAPER

P. A. Mothes 7 M. L. Hall 7 R. J. Janda

The enormous Chillos Valley Lahar: an ash-flow-generated debris flow

from Cotopaxi Volcano, Ecuador

Received: 28 April 1997 / Accepted: 19 August 1997

Abstract The Chillos Valley Lahar (CVL), the largest

Holocene debris flow in area and volume as yet recognized in the northern Andes, formed on Cotopaxi volcanos north and northeast slopes and descended river
systems that took it 326 km northnorthwest to the Pacific Ocean and 130c km east into the Amazon basin.
In the Chillos Valley, 40 km downstream from the volcano, depths of 80160 m and valley cross sections up
to 337 000 m 2 are observed, implying peak flow discharges of 2.66.0 million m 3/s. The overall volume of
the CVL is estimated to be ;3.8 km 3. The CVL was
generated approximately 4500 years BP by a rhyolitic
ash flow that followed a small sector collapse on the
north and northeast sides of Cotopaxi, which melted
part of the volcanos icecap and transformed rapidly
into the debris flow. The ash flow and resulting CVL
have identical components, except for foreign fragments picked up along the flow path. Juvenile materials, including vitric ash, crystals, and pumice, comprise
8090% of the lahars deposit, whereas rhyolitic, dacitic, and andesitic lithics make up the remainder. The
sand-size fraction and the 2- to 10-mm fraction together
dominate the deposit, constituting ;63 and ;15 wt.%
of the matrix, respectively, whereas the silt-size fraction
averages less than ;10 wt.% and the clay-size fraction
less than 0.5 wt.%. Along the 326-km runout, these particle-size fractions vary little, as does the sorting coefficient (averagep2.6). There is no tendency toward
grading or improved sorting. Limited bulking is recognized. The CVL was an enormous non-cohesive debris
flow, notable for its ash-flow origin and immense volume and peak discharge which gave it characteristics
Editorial responsibility: D. A. Swanson
Patricia A. Mothes (Y) 7 Minard L. Hall
Instituto Geofsico, Escuela Politcnica Nacional, Quito
Ecuador
Fax: c593 2 567847
e-mail: instgeof6uio.satnet.net
Richard J. Janda
U.S. Geological Survey, Vancouver, Washington, USA

and a behavior akin to large cohesive mudflows. Significantly, then, ash-flow-generated debris flows can also
achieve large volumes and cover great areas; thus, they
can conceivably affect large populated regions far from
their source. Especially dangerous, therefore, are snowclad volcanoes with recent silicic ash-flow histories such
as those found in the Andes and Alaska.
Key words Debris flow 7 Pyroclastic flow 7 Cotopaxi 7
Ecuador 7 North Andes

Introduction
The eruptions of Mount St Helens (MSH) in 1980, Nevado del Ruiz (NR) in 1985, Redoubt in 19891990 and
Pinatubo in 1991 resulted in many in-depth studies of
debris flows (MSH and Redoubt: Janda et al. 1981,
Pierson 1985, Scott 1988, Waitt et al. 1983, Trabant et
al. 1994; NR, Colombia: Pierson et al. 1990; Pinatubo:
Pierson et al. 1992). These flows had matrices which
were mainly fines-poor and non-cohesive, important
factors which contributed to their miscibility with river
water and eventually aided their transformation to hyperconcentrated streamflows (Scott 1988).
Much less common are debris flows that have a cohesive matrix with clay-size particle contents greater
than 35% and that travel great distances without significant textural variation and with little attenuation of
their peak discharges (Scott et al. 1995). Lahars of this
nature are exemplified by the 3.8 km 3 Osceola mudflow, generated some 5600 years BP by a sector collapse of Mount Rainier (Crandell 1971; Vallance and
Scott 1997), which experienced few textural changes
along its runout. Given the large discharges that were
apparently sustained for great distances by these flows,
their generation poses significant hazard for population
centers downstream.
Debris flow formation requires (a) a significant water source (ice/snow cover, intense rainfall, crater lake,
etc.), (b) abundant unconsolidated material (such as
tephra, pyroclastic flow deposits, glacial outwash), (c)

234

great relief and steep slopes on the cone, and (d) a triggering mechanism (Major and Newhall 1989). The volume of the flow is influenced by the above factors and
by the nature of the eruptive activity. For example,
fluidized pyroclastic flows of silicic composition,
charged with fine-ash particles, are capable of efficient
heat transfer as well as vigorous erosion and intermixing with ice and snow, thereby accelerating the conversion of ice to water (Pierson et al. 1990; Pierson
1995).
At Cotopaxi Volcano, the debris flows with the
largest volumes are those that formed by interaction
between silicic pyroclastic flows and glaciers. This article describes one of the largest Holocene lahars yet reported, formed by the instantaneous melting of Cotopaxis glacial cap by a pyroclastic flow of rhyolitic composition. This the Chillos Valley Lahar (CVL) had
an ash-rich matrix which allowed it to travel 326 km
from its source without undergoing significant textural
transformation, a characteristic of cohesive debris
flows. Similar rheological behavior occurred in the Osceola mudflow, which flowed 120 km without important
textural changes (Crandell 1971), attributed mainly to
its high clay content. The CVL, however, having few
clay-size particles and yet sharing many of Osceolas
characteristics, probably owes its behavior to its immense volume.

Fig. 1 Cotopaxis north flank and possible avalanche scar that

heads at the rock face near its 5890 m summit. Relief is 2100 m
over its 3800-m-high base. In middle ground are debris flow channels occupied by historic lahar deposits

Cotopaxi Volcano
Cotopaxi Volcano (5890 m), a large symmetrical cone
with 3000 m of relief, is one of the principal stratovolcanoes of Ecuadors eastern volcanic row, overlying a metamorphic basement of approximately 3000 m elevation
(Fig. 1). Its upper glacier-clad flanks have slopes averaging 25307, whereas its lower flanks are incised by canyons tens of meters deep. Its snow and ice fields cover
approximately 20 km 2 with a volume of 1.0 km 3 (Jordan 1983). Like elsewhere in the Andes, these glaciers
are presently in retreat.
Three important drainages head on the volcano: the
Rios Pita-Esmeraldas to the north, the Rio CutuchiPastaza system to the southwest, and the Rio TamboTamboyacu-Napo system to the east (Fig. 2). The latter
two drainages ultimately descend the eastern flank of
the Andes to join the greater Rio Amazon system. The
Rio Pita flows northward approximately 90 km through
the Inter-Andean Valley, then descends the western
flank of the Andes and subsequently enters the Pacific
Ocean at the city of Esmeraldas, after a 326-km journey. To the southwest, the Rio Cutuchi flows southward down the Inter-Andean Valley for 130 km and
then eastward through the Cordillera Oriental to become an Amazon tributary.
The Rios Pita and Cutuchi both flow through the 50km-wide Inter-Andean Valley, home to approximately
3 million inhabitants and numerous towns and cities,
some of which have been severely affected by Coto-

Fig. 2 Partial map of Ecuador showing the Inter-Andean Valley,

bordered by the Cordillera Occidental and the Cordillera Oriental, and occupied by Cotopaxi Volcano and the adjacent Chillos
and Latacunga valleys. The Ro Pita-San Pedro-GuayllabambaEsmeraldas river system enters the Pacific Ocean near the city of
Esmeraldas. The Ros Napo and Pastaza join the Amazon River
downstream

paxis lahars on several occasions since the arrival of

the Spanish in 1532 (Mothes 1992).
Cotopaxi has had approximately 30 eruptions since
1532 (Hall 1977). These eruptions were typically accompanied by regional scoria and pumice ash falls,

235

blocky lava flows, scoria pyroclastic flows, and lahars,

all of andesitic character (SiO2p5658%). However,
during the past 10 000 years, rhyolitic eruptions have
periodically taken place in its predominantly andesitic
development. At least four such eruptive periods have
occurred at intervals of approximately 2000 years, producing moderate-size ash flows (Hall and Mothes
1995).
Around Cotopaxis base several cream-colored, ashrich debris-flow deposits are recognized that are primarily comprised of fine-grained rhyolitic ash and are notable for their large volumes and areas. They can be
readily traced upslope, where they become indistinguishable from rhyolitic ash flows. The largest of these deposits in Cotopaxis recent past is the Chillos Valley
Lahar, formed when the north and northeast sides of
the cone suffered a sector collapse and moderate-size
Fig. 3 Map of Chillos Valley
and the Ro Pita-San Pedro
river system. Limits of the
Chillos Valley Lahar (CVL)
deposit are shown, as well as
representative avalanche hummocks and the Canon Colorado ash-flow deposit. Note
the selected channel cross sections of the CVL unit. Sample
locations refer to granulometric analyses

ash flows were generated. This debris flow descended

primarily the Rio Pita passing through the Chillos Valley (Fig. 3), but it also gained access to the northern tributaries of the Rio Cutuchi and descended that drainage for at least 130 km (Fig. 4).

Characteristics of the Chillos Valley Lahar

The CVLs passage through Ecuadors Inter-Andean
Valley was probably one of the most catastrophic geologic events to have occurred during the Holocene. The
extensive area that the lahar covered, its huge volume,
its beige color and unique composition, and the volcanic events associated with its generation distinguish it
from other Cotopaxi debris flows. In the Chillos Valley,
located 20 km southeast and 400 m lower than Quito,

236

Fig. 4 Map of Latacunga Valley and the major streams draining

the southwest side of Cotopaxi Volcano to form the Ro Cutuchi.
Limits of the CVL deposit are shown. Neither avalanche nor ashflow deposits are found on the west side of the volcano. Sample
locations refer to granulometric analyses

the flows path was more than 11 km wide and had a

depth of 150 m. This prehistoric event also had its human toll, although of unknown proportions, as sherds
of prehistoric pottery are found at the base of the flow
or in the underlying soil.
Flow routes and area
The CVL inundated areas substantially greater than
the narrow valleys occupied by more recent, smaller lahars generated at Cotopaxi. From historical accounts
and stratigraphic studies, it is known that the majority
of the small debris flows since 1532 flowed down the
west and southwest tributaries of the Rio Cutuchi,

whereas few followed the northern drainages. However, in the case of the CVL, its larger portion descended
Cotopaxis northern rivers, where it is seen in cuts in
the broad plain at the volcanos northern base (Limpiopungo), downvalley along the Rios San Pedro and Pita,
and in the Chillos Valley. The CVL also rode up onto
the lower flanks of both Ruminahui and Sincholagua
volcanoes. To the west its route was restricted to the
Rios San Lorenzo and Cutuchi. All channels have high
gradients (45%) for the first 2540 km and lower gradients (12%) thereafter, until they descend the Andes
steep flanks (Fig. 5). As compared with the heights and
typical stream gradients of other major lahar-generating volcanoes, Cotopaxi is distinguished by its extreme
relief (;6000 m) and two steep descents to reach sea
level, that of the cone itself (;3500 m drop) and that of
the Andes western flank (;2500 m drop).
Along the northern flow route, approximately 60%
of the discharge was diverted westward to the Rio San
Pedro by Pasochoa Volcano (A-A in Fig. 3). The remaining ;40% of the CVL continued down the Rio
Pita channel, where the flow was constricted and attained depths up to 200 m (B-B in Fig. 3). Upon leaving
the confines of the San Pedro and Pita canyons, with
reaches of approximately 25 and 15 km, respectively,
the debris flow spread laterally into the Chillos Valley.
Because of its shorter path, the Rio Pita flow probably
arrived first in the Chillos Valley, possibly creating a
temporary barrier for the San Pedro flow which arrived
slightly later.
Once the two flows merged, a formidable cross-sectional width of 11 km was attained. Farther downvalley
near Conocoto the width diminished to approximately
6 km, and well-defined limits show that depths greater
than 90 m were achieved (D-D in Fig. 3). For the next
12 km downstream, cross-sectional widths first decreased as the flow was confined to the San Pedro canyon and then increased as it spread into the CumbayTumbaco Valley, where depths up to 120 m are observed (E-E in Fig. 3). To this point the CVL had covered approximately 270 km 2.
Downstream from Cumbay the CVL followed the
narrow and steep-walled canyon of the Rio Guayllabamba for 186 km, along which gradients range from
16% (Fig. 5). Little deposition occurred along this section due to the high-energy environment. However, the
CVL deposit is found interbedded with fluvial sediments over the last 75 km of the Rio Esmeraldas floodplain until it reaches the Pacific Ocean (Fig. 2). Thus,
this debris flow traveled 326 km and left a deposit, averaging 2 m thick, over 440 km 2. Its submarine extension onto the offshore platform is unknown.
The CVL deposit is readily identified in outcrops
west of Cotopaxi, thanks to its high stratigraphic position and exceptional lithologic and textural characteristics. Upon leaving the high-gradient narrow canyons of
the Rios Cutuchi and San Lorenzo, the flow spread
downstream to widths of 35 km for the next 25 km
along the Latacunga Valley (Fig. 4). The flow heights

237

Fig. 5 Longitudinal profiles of the two principal river channels

draining Cotopaxi Volcano. Note horizontal scale change. River
profiles at Nevado del Ruiz, Mount St. Helens, and Redoubt volcanoes are shown for comparison. (Data from Pierson et al. 1990,
Pierson 1985, and Trabant et al. 1994, respectively)

did not exceed 20 m above the valley bottom; thus,

more recent lahars often cover the CVL and consequently outcrops are few. Where channel gradients are
0.5% or less, deposits tend to have thicknesses of
23 m. Between Latacunga and Panzaleo, a distance of
17 km, the flow was constricted to a 2- to 3-km-wide
path and was 3540 m deep. South of Panzaleo the flow
entered a steep-walled canyon, in which few deposits
are preserved. Approximately 160 km 2 were flooded by
the debris flow in the Latacunga Valley downstream to
Panzaleo.
Volume and discharge of the CVL
Good preservation of the CVL deposit allows its volume to be estimated. Profiles based upon well-defined
strandlines left by the flow at three sites in the Chillos
and Cumbay valleys give cross-sectional areas of
337 000 m 2 at Conocoto (45 km from the summit),
130 000 m 2 at Guangopolo (50 km from the summit),
and 138 000 m 2 at Cumbay (62 km from the summit).
For the Latacunga Valley, areas of 102 000 m 2 at Latacunga and 71 000 m 2 at Salcedo are calculated (Figs. 3,
4).
Field evidence for the velocity of the CVL is lacking.
Velocity estimates for Cotopaxis clast-rich 1877 debris
flow obtained from runup calculations and numerical
modeling give minimum speeds of ;57 m/s in the
Chillos Valley (Castro et al. 1992). However, large-volume debris flows are thought to have substantially
greater velocities. An empirical study of the travel
timedistance relationships for debris flows of varying
size suggests that flows with peak discharges
61!10 6 m 3/s probably travel several orders of magni-

tude faster than medium-size flows (Pierson, in press).

The large cohesive Osceola mudflow is estimated to
have had a velocity of ;19 m/s at 4050 km downstream from its source (Vallance and Scott 1997). If the
CVL had a velocity similar to that of the Osceola Mudflow, a debris flow comparable in size and behavior to
the CVL, peak flow discharges in the Chillos Valley
could have been of the order of 2.66.0!10 6 m 3/s, values far exceeding the daily discharge of the present
Amazon River.
Along the northern drainage the CVL deposit has a
minimum total dry volume of approximately 0.90 km 3,
which is calculated from an average 2-m-thick deposit
that mantles 440 km 2. Along the southwest drainage, a
minimum total dry volume of 0.32 km 3 is estimated. We
assume that the CVL was composed of ;30% water
and ;70% sediment by volume the approximate ratio reported for lahars at Mount St. Helens (Pierson
and Scott 1985) and other debris flows (Sharp and Nobles 1953; Pierson 1980), as well as by our experiments
that show that a 28% volume of water is needed to initiate mobility of the dry CVL material. As such, the
CVLs water-saturated volume would have been at
least 1.7 km 3, a minimum value which does not consider the volume that entered the sea and the Amazon basin.
As discussed later, the CVL is thought to have transformed completely from an ash flow, whose volume is
estimated at 2.5 km 3. This implies a total wet volume of
approximately 3.8 km 3, which does not consider the
volume of material picked up along the runouts.
Textural and compositional characteristics
The CVL deposit is readily identified by its beige-tan
color, its homogeneous appearance, and the fact that it
is often the topmost unit and mantles the countryside
(Fig. 6). Equally important, in the field it appears to
consist of ;8090% matrix, composed of ash, pumice,
and lithic grains, as well as ;1020% lithic clasts larger

238

Fig. 6 Typical CVL outcrop, here overlying a blocky-structured

soil. Note the CVLs matrix-rich character, homogeneous distribution of clasts, and absence of any stratification. Taken at San
Rafael, 45 km downstream from Cotopaxi

than 12 cm. In a few places dense boulders up to 2 m

in diameter occur. Throughout the Chillos and Tumbaco valleys, and even along the Rio Esmeraldas in the
coastal zone, the deposit has a uniform thickness of approximately 2 m. Thicknesses of 34 m are attained in
paleochannels. Lateral pinching out of the deposit occurs at the flow margins.
The CVL deposit is distinguished from the ash flow
by containing significantly more lithic clasts (up to 30 vs
8% by area), carrying rounded pebbles and cobbles of

dense rocks, having a greater variety of polylithic fragments many of which are accidental, as well as having a
harder, more compacted matrix.
No internal stratification and only rare segregation
of larger clasts are observed. Furthermore, there is no
evidence of individual flow pulses or basal sole layers
often reported in debris flow deposits. The flow was
erosive, as shown by the occasional incorporation of
soil clots, rounded stream cobbles, and the presence of
linear casts of the CVLs basal material corresponding
to grooves cut into the underlying soil by dragged
stones. The CVL matrix contains small microvesicles,
remnants of entrapped air and gas in the original matrix (Fisher and Schmincke 1984).
Clast and matrix characteristics were determined for
representative samples of the deposit along both the
northern and southern flow paths. Clast parameters
were obtained by simple counting and classification as
to size and type of fragments directly on the outcrop
face. As seen in Table 1 the population of clasts greater
than 10 mm in diameter accounts for ;1867% of the
surface area, averaging ;2530%. Thus, the matrix
(^10-mm-size grains) varies from ;3382%, averaging ;7080% of the surface area. In the ash-flow tuff
associated with the CVL, the clasts (610 mm) and matrix comprise 8 and 92% of the outcrop face, respectively.
Granulometric analyses of the ^64 mm size fraction
of the deposits were carried out by routine sieve and
hydrometer methods (Table 2). Sand-size grains
(0.0632 mm) greatly dominate the matrix (4779 wt.%;
Fig. 7). The 2- to 10-mm size fraction varies from
844 wt.%, whereas the ^0.063-mm size fraction
ranges from 121 wt.%. The clay-size particle content
of five samples as determined by hydrometer analysis
remains at only 0.240.41% of the matrix.
Noteworthy is the consistent size distribution with
distance, even over the exceptionally long runout of the
CVL (Figs. 7, 8). The sand-size fraction of most samples has a limited range of 5967%, similar to the 61%
shown by the ash flow (sample 1; Table 2). The 2- to
10-mm fraction also shows a limited variation (819%),
with few exceptions. Some samples (nos. 2 and 6) have

Table 1 Size distribution of clasts

Sample Location

Distance
Area of
Clast count of area %)
downstream (km) outcrop (m 2)
1 128 mm 64128 mm 3264 mm 1032 mm Total
Clast/m 2 Matrix/m 2

Northern route
PF-campsite
1A
CVL-campsite
1B
Pedregal
6
Selva Alegre
11
Conocoto
12
Tumbaco
13
Viche
14

7
7
20
42
48
65
270

1.70
0.80
3.90
3.74
3.90
2.70
4.06

0.00
15.09
6.16
2.14
5.13
8.87
61.08

1.09
6.96
12.31
4.93
3.31
6.13
3.63

0.54
2.60
7.20
5.47
5.96
2.81
1.59

6.16
22.81
2.97
5.26
11.27
13.00
0.28

7.79
47.46
28.64
17.80
35.67
30.81
66.58

92.21
52.54
71.36
82.20
74.33
69.19
33.42

239
Table 2 Granulometric analysis: Chillos Valley Lahar
Sample
no.

Location

Distance
from
crater (km)

Unit
thickness
(m)

Position
in
deposits

Content (wt.%)
Gravel
6410 210
mm
mm

1
2
3
4
5
6
7
8
9
10
11
12
13
14

N. Route, Chillos Valley

7
Campsite PF
9
Reservoir Av
11
Bulldozer
14
N. Park entrance
18
Pedregal
20
Pedregal/Av
23
La Caldera
30
Machachi
33
Hushpuppy
41
Chillo Jijon
42
Selva Alegre
48
Conocoto
65
Tumbaco
270
Viche

15
16
17
18

S. Route, Latacunga
Cutuchi Pf
Cutuchi CVL
Latacunga
Nagsiche

Valley
15
15
43
56

Sand

Silt
and
clay

10.0
5.0
2.0
2.5
2.0
6.0
3.0
2.5
2.0
2.5
1.5
3.0
3.0
2.0

t/c
c
c
c
b
b
c
c
c
c
c
t/c
c
t/c

1.4
7.8
2.3
5.2
6.9
14.7
5.4
14.0
13.7
17.0
3.8
1.8
7.3
5.4

17.5
43.9
14.8
26.0
7.8
35.7
10.0
18.6
19.0
9.0
19.2
11.8
13.5
19.1

61.0
46.9
61.5
60.2
78.5
48.2
76.6
61.1
64.5
59.0
63.6
75.8
67.5
65.4

20.1
1.4
21.4
8.6
6.4
1.4
7.8
6.3
2.8
15.0
13.0
10.5
11.4
9.8

3.0
3.0
2.0
4.0

c
c
t/c
t/c

4.3
5.9
6.5
2.3

30.1
13.2
15.5
13.8

65.1
67.1
66.0
71.9

0.5
13.8
12.0
12.0

a greater amount of coarser material, reflecting their

avalanche debris contents. The mean grain size for all
samples is 0.70 mm with a range of 0.251.80 mm
(Fig. 9A). As compared with the cohesive Osceola and
Electron debris flows (Fig. 7), the CVL deposit is clearly dominated by the sand-size fraction and contains
fewer coarse-grained clasts as well as less silt and claysize material (Fig. 9C). Its sorting coefficients (sG)
have a mean value of 2.6 within a variation of 2.03.7
(Fig. 9B); the CVL is slightly better sorted than the others. The CVL deposit, like the Osceola deposit, appears
to preserve much of the granulometric character of the
original source material, despite its transformation
from a pyroclastic flow and the great distance traveled.
The clay-size fraction of the CVL is negligible compared with that of debris flows formed by collapse of
hydrothermally altered edifices, such as at Mount Rainier (Scott et al 1995; Vallance and Scott 1997) or at Citlaltpetl volcano, Mexico (Carrasco-Nnez et al. 1993).
Cotopaxi shows no evidence of intense alteration in its
recent history. In fact, the CVL material is composed
almost exclusively of fresh, juvenile pyroclastic products. Even compared with non-cohesive lahars, such as
those of Pine Creek at Mount St. Helens and of the Rio
Azufrado at Nevado del Ruiz, the CVL deposit contains fewer clay-size particles. In Fig. 10 cumulative
grain size frequency curves for the CVL are compared
with curves for the cohesive Electron and Osceola lahars and the non-cohesive Molino (NR) lahar; the
curves again emphasize the dominant sand-size fraction, the scarce silt-and-finer fraction, and the slightly
better sorting of the CVL deposit.

Clay

0.41
0.24

0.40
0.35
0.31

Mean Grain
Size (Mz)

Sorting

phi

mm

oG

1.65
P0.40
1.55
0.37
1.03
P0.83
1.30
0.36
0.03
0.67
1.20
2.00
1.10
1.07

0.32
1.35
0.34
0.80
0.50
1.80
0.41
0.82
0.99
0.65
0.46
0.25
0.48
0.48

2.8
2.5
2.6
2.8
2.0
2.6
2.3
2.9
2.8
3.7
2.7
2.4
2.7
2.8

0.00
0.87
1.03
1.37

1.00
0.55
0.49
0.40

2.0
2.1
2.7
2.3

Nine major clast types are common in the CVL unit:

gray rhyolitic vitrophyre, red- and black-banded rhyolite, perlitic obsidian, gray andesitic blocks, white microvesicular biotite-bearing pumice, scoria lapilli and
bombs, rounded stream cobbles, clods of the underlying soil and sediments, and uncommon pods of gray andesitic and rhyolitic breccia.
The obsidian fragments, gray rhyolite, pumice, and
banded rhyolite are identical to the lithic clasts in the
pyroclastic flows and tephra falls of the associated Canon Colorado ash flow sequence discussed below. The
breccias and andesitic blocks formed the pre-eruption
cone and were caught up in the lahar. The rhyolite, obsidian, and pods of rhyolitic breccia are remnants of
pre-existing domes. Rounded oxidized cobbles are frequently found and were incorporated from the stream
bed. Clods of soil and indurated tuff and pods of loose
pumice, ranging from 1 cm to 2 m in diameter, are observed in the deposits but amount to ^1% of the volume.
Along the entire 326 km course of the CVL, except
in a few areas, there is no tendency toward grading or
improved sorting (Fig. 9B). Dense lithic clasts up to 2 m
in size are found in the deposit in a few places and
show no tendency to concentrate at the base or the top
of the flow. This aspect reflects the high-yield strength
of the flow along its entire course, a characteristic of
well-developed debris flows (T. C. Pierson, pers. commun.).
That the CVL unit maintains its integrity with distance is noteworthy, given its long path down high-gradient channels, including the Rio Guayllabamba canyon, where one would expect mixing and dilution with

240

Fig. 8 Cumulative grain-size frequency curves for CVL samples

(^64 mm) along the northern runout for flow margin and flow
axis samples. The curves are almost identical in shape and position for the entire runout, with the exception of the Conocoto
sample which is from the extreme western margin of the flow

river water and transformation to a hyperconcentrated

flow. However, there is no separation or segregation of
coarser or denser clasts, and the matrix that arrived at
the Pacific Ocean is seemingly identical to the initial facies; these two characteristics strongly imply that little
interaction with river water took place. This may have
been due to the CVLs limited opportunity to mix with
river water given its presumed viscous nature and rapid
passage. Scott et al. (1995) found that a sand-rich matrix tends to retard dilution of a flow by stream water as
well as the settling out of coarse clasts. Additionally,
the flow front may not have been greatly influenced by
stream and river water encountered along the runout,
which presently constitutes a relatively minor quantity
as compared with the CVLs immense volume.
Other aspects of the CVLs unusual character

Fig. 7 Grain-size distribution of the CVL matrix (particles

^64 mm) along northern runout. Similar graphs are shown for
the Osceola and Electron debris flows from Mount Rainier Volcano. Data from Crandell (1971). Note the dominance of finegrained material in the CVL unit

Electro-chemical attractions inherent to clay mineral

particles have been cited to explain the cohesiveness of
fine-grain mudflows (Qian et al. 1980; Rodine 1974;
Major and Pierson 1992). The important yield strength
of the CVL during its long journey cannot, however, be
attributed to clay minerals, since their quantity is negligible (Table 2). Here it is suggested that the high concentration of irregularly shaped particles of volcanic
glass resulted in a relatively tight packing of the CVLs
matrix that would have inhibited both mixing with water and the winnowing of the fine fraction, as well as
contributed to the CVLs notable yield strength.
Bulking was a factor in the CVLs behavior. As
mentioned, pods of soil and pre-existing tuff are scattered throughout the deposit. A plot of clast type vs
distance (Fig. 11) shows the relative proportions of foreign fragments incorporated from the stream bed as
compared with andesite, rhyolite, and pumice clasts
carried from Cotopaxi. Andesite and other rock types

241

Fig. 10 Cumulative grain-size frequency curves for three CVL

samples taken 33, 42, and 270 km from source as compared with
the Electron and Osceola debris flows, Mount Rainier (data from
Crandell 1971), and the Molinos lahar, Nevado del Ruiz (data
from Pierson et al 1990). The CVL unit demonstrates slightly better sorting than the other flows, reflecting homogeneity of original ash-flow material

Fig. 9 Downstream variation in A mean grain size [graphic mean

(Mz) in mm], B sorting coefficient [graphic standard deviation
(sG)], and C percentage of silt and clay-size particles in matrix
along the northern runout (in wt.%). In Fig. 9A the relatively
constant trend is perturbed by samples 2 and 6 which carry coarser avalanche material. Grain-size parameters from Folk (1980)

were picked up as the flow descended the steep canyons leaving the Andes and were then deposited upon
the upper coastal fan. The loss of pumice clasts was
probably due to abrasion rather than winnowing, given
their fast attrition rates and the long flow path. The
large amount of incorporated rhyolitic material observed at the 20-km site is probably related to nearby
avalanche hummocks.
Flow bulking is also implied by a plot of the number
of rounded and subrounded, dense (non-pumice) clasts
per square meter of outcrop vs distance, although a
clear progressive increase with distance is not indicated.
Figure 12 shows a relatively constant number of these
clasts with distance, especially for the 632 mm size.

Fig. 11 Percentage of the four principal clast types that comprise

the clast population (Table 1) found in the CVL deposit vs distance from source. All samples from the Ro Pita-Esmeraldas runout. Size of clasts 610 mm. Pum/Scor Pumice and scoria; Rhy/
Dac rhyolite and dacite

Apparently, many clasts settled out while crossing the

wide coastal plain. The sharp increase of rounded, 8- to
32-mm-size clasts observed at the 55-km site (Fig. 12) is
apparently related to its close proximity to the main
channel.

Origin of the Chillos Valley Lahar

The sector collapse and pyroclastic flow events
In Canon Colorado at the northern foot of Cotopaxi,
three sequential pumice-rich rhyolitic ash flow units are
exposed, which are bounded upward and downward

242

Fig. 12 Number of rounded and subrounded dense clasts per

square meter of CVL outcrop vs distance from source. All samples from the Ro Pita-Esmeraldas runout. The large number of
rounded clasts observed at the 55-km site may be related to its
proximity to the main channel

stratigraphically by typical andesitic deposits of this

volcano (Hall and Mothes 1995). A few kilometers
downvalley the deposits of the larger ash flows, here
called the Canon Colorado pyroclastic flows (CCPF I
and II), are associated with clast-rich hummocks and
overlain intimately by the CVL unit (Fig. 13). At these
outcrops and others around the northeast foot of the
cone, the avalanche deposits that make up these hummocks are composed primarily of angular clasts of obsidian, banded rhyolite, gray aphyric rhyolite, and dacite in a sandy matrix of the same lithologies, although
older andesitic clasts are also found. They are attributed to a small sector collapse of the upper north and
northeast flank of the cone which included older rhyolitic domes, remnants of which are presumed to underlie
the young tephra cover of the volcano. The avalanche
deposit is found distributed around the north and
northeast foot of the cone and downvalley for 25 km,
which suggests an overall volume of approximately
2 km 3.
Field mapping shows the lower ash flow deposit to
be the most widespread and voluminous of the units.
The main trajectory of this flow was northward and
northeastward, where it covered the Limpiopungo
plain and the Rio Pita Valley with a deposit tens of meters thick (Fig. 3). We estimate that the deposit once
covered ;170 km 2 and had a volume of ;2.5 km 3.
Both the CCPF-I and -II units are matrix-rich ash
flow tuffs containing ;92% vitric ash, crystals, and pumice particles that comprise the matrix (^10 mm), as
well as ;8% clasts (610 mm) of microvesicular pumice, obsidian, and black, red, and gray-banded aphyric
rhyolite. The pumice typically contains 12% gold and
black biotite, 510% quartz and plagioclase, 2% magnetite, and a trace of hyperstene and K-feldspar. Both
the mineralogy of the pumice and the lithic clasts of the
CCPF are identical to those found in the CVL deposit.
The SiO2 contents of the pumice from these units are
7173% (M. L. Hall and P. Mothes, unpublished data).

Fig. 13 Interrelations of the CCPF and the overriding avalanche

debris followed by the CVL. Photo taken at the northern base of
the cone, west of sample site 2 (see Fig. 3)

Consequently, the overall composition of the CVLs

matrix (i.e., 8090% vitric ash, pumice, and crystals,
and 1020% of small fragments of dense pumice and
rhyolite) accords well with that of the CCPF.
The deposits of the Chillos Valley Lahar lie in intimate association with those of the hummocks and the
CCPF; the CVL incorporated pods of loose avalanche
breccia and of the lower and middle ash flows. At many
outcrops varied and reversed stratigraphic relations exist between the CVL, the ash flows, and the avalanche
breccia, suggesting that all three events were approximately simultaneous.
Sources of water
Glacial studies in the Ecuadorian Andes suggest that at
approximately 5000 years BP glacier limits on volcanoes were 300500 m lower than their present limits of
4800 m elevation (Clapperton 1990). Thus, at the time
of the CVLs formation the glacial ice cap on the north,
northeast, and northwest quadrants of Cotopaxis cone
may have had a total area of approximately 20B5 km 2,
corresponding to a volume of approximately 1.01.3
km 3 (assuming an ice thickness of 3050 m). The maximum amount of water available, if all of this ice melted,
would have been approximately 0.81 km 3, sufficient to
have mobilized approximately 2 km 3 of dry debris.
Additional sources of water include ground water
from the volcano, possible moraine-dammed lakes, and
river water. Undoubtedly, the pre-collapse cone contained substantial water, for it was glacier clad. If the

243

volume of the sector collapse was of the order of 2 km 3,

as postulated previously, and contained approximately
5% ground water, an additional 0.1 km 3 of water would
have been available. Small moraine-dammed lakes are
not uncommon on present-day volcanoes and could
have contributed water. However, given that morainal
debris is not observed in the proximal CVL deposits,
this source is discounted. As discussed previously, the
amount of water encountered downstream along the
runout was probably not large (estimated at
0.0075 km 3) as compared with the immense volume of
the CVL, especially for the first 250 km. Thus, rivers
probably contributed little to the overall water required.
Age of the CVL
The lahars age is given by several radiocarbon dates.
The soil beneath the Canon Colorado ash flows on Cotopaxis north flank is dated at 4670B70 14C years BP
(Smyth 1991). Tree trunks taken from the CVL deposit
on the coastal plain give two dates of 4500 14C years BP
(N. Banks, pers. commun.). Lastly, pottery and figurines of black coral found directly beneath the unit are
assigned a mid-Valdivian age, thought to be approximately 4500 years old (P. Norton, pers. commun.).
Based on these dates as well as on the logical succession of younger dates for the overlying deposits, we
conclude that the CVL was most likely generated between 4500 and 4600 years ago (uncorrected).

Conclusions
The Chillos Valley Lahar, a debris flow of immense
proportions, was generated in a sequence of events that
began with the collapse of the north and northeast quaFig. 14 Probable sequence of
events leading to the formation of the CVL. Downvalley
transit is to the north

drants of Cotopaxi Volcano (Fig. 14), involving parts of

the andesitic edifice but mainly rhyolitic domes, the
likely remnants of the previous rhyolitic activity that
had occurred at ;6000 years BP (Hall and Mothes
1995). The resulting avalanche left a hummocky plain
at the north and northeast base of the cone. More importantly, the collapse probably thoroughly fragmented
the existing glacial icecap, a phenomenon previously reported at Mount St. Helens and Nevada del Ruiz volcanoes (Waitt et al. 1983; Pierson et al. 1990). Immediately afterward, the CCPF erupted from the site of the collapse, rapidly melted much of this fractured ice and
snow (reported elsewhere by Pierson and Janda 1994;
Pierson 1995), became saturated with water, and transformed quickly into the CVL. Because only a single
flow unit is observed in either the proximal or distal
reaches, it appears that once this immense debris flow
was generated, little water remained available for additional debris flow generation. Whereas the ash flow and
CVL were mainly directed down the Rio Pita and San
Pedro drainages, a significant portion of the CVL was
diverted westward by the drainage divide at the northern foot of the cone and thus gained access to the Latacunga Valley. It probably flowed rapidly through the
Inter-Andean Valley, both northward and southward,
in a few hours.
Apparently the ash-rich pyroclastic flow transformed immediately and almost entirely into the CVL
near the base of the cone. With few exceptions, both
share an identical mineralogical and lithological composition and grain-size distribution.
As such, the CVL was a unique non-cohesive debris
flow, exceptional because of its ash-flow origin and
enormous volume, as well as having behavioral characteristics similar to those of large cohesive mudflows.
The flows great volume and discharge played a fundamental role, since it apparently prevented the potential
dilution of the debris flow by river water, which in turn

244

inhibited significant textural changes with distance, despite the very long 326-km runout.
At Cotopaxi, this type of debris flow has a low recurrence rate compared with the more common andesitic debris flows generated many times each century.
Nevertheless, given that rhyolitic eruptions have taken
place on Cotopaxi every 15002000 years during the
Holocene, other CVL-like debris flows may happen
again.
Recent studies of large debris flows generated on
hydrothermally altered stratovolcanoes (Vallance and
Scott 1997) emphasize the potential danger of these
events, given their volumes, speeds, and ability to
spread out over large areas. The present study demonstrates that ash-flow-generated lahars can also attain
impressive flow dimensions. Consequently, pristine volcanoes with silicic ash-flow histories and large water resources, such as those in the Andes and Alaska, should
be considered potential candidates for generating very
large debris flows that can cause widespread destruction far from their sources.
Acknowledgements We especially thank our deceased friend
and co-author, Dick Janda, who over the years shared his knowledge and enthusiasm with all of us who work on lahar problems.
We also acknowledge the help of T. Pierson and K. Scott, who
reviewed a previous version of the manuscript, and J. Beget and
G. Smith for the final draft. We also thank the on-going support
of the VDAP program of the U.S. Geological Survey and its
members.

References
Carrasco-Nnez G, Vallance JW, Rose WI (1993) A voluminous
avalanche-induced lahar from Citlaltpetl Volcano, Mexico:
implications for hazard assessment. J Volcanol Geotherm Res
59 : 3546
Castro M, Mothes P, Hidalgo X, Snchez A (1992) Uso del Modelo HEC-1 para el estudio de los lahares del Volcn Cotopaxi.
Politcnica 17 : 163178
Clapperton CM (1990) Quaternary glaciations in the southern
hemisphere: an overview. Q Sci Rev 9 : 299304
Crandell DR (1971) Postglacial lahars from Mount Rainier Volcano, Washington. Geol Surv Prof Pap 677 : 173
Fisher RV, Schmincke HU (1984) Pyroclastic rocks. Springer,
Berlin Heidelberg New York
Folk RL (1980) Petrology of sedimentary rocks. Hemphill, Austin, Texas
Hall ML (1977) El Volcanismo en el Ecuador. Instituto Panamericano de Geografa e Historia, Quito
Hall ML, Mothes P (1995) Bimodal nature of the eruptive history
of Cotopaxi Volcano, Ecuador. Abstract IUGG XXI General
Assembly, Boulder, Colorado, p A452
Janda RJ, Scott KM, Nolan KM, Martinson HA (1981) Lahar
movement, effects, and deposits. In: Lipmen PW, Mullineaux
DR (eds) The 1980 eruptions of Mount St. Helens, Washington. US Geol Surv Prof Pap 1250 : 461478
Jordan E (1983) Die Vergletscherung del Cotopaxi, Ecuador.
Gletscherkunde und Glaziologie 19 : 73102
Major JJ, Newhall CG (1989) Snow and ice perturbation during
historical volcanic eruptions and the formation of lahars and
floods. Bull Volcanol 52 : 127

Major JJ, Pierson TC (1992) Debris flow rheology: experimental

analysis of fine-grained slurries. Water Resour Res
38 : 841857
Mothes PA (1992) Lahars of Cotopaxi Volcano, Ecuador: hazard
and risk evaluation. In: McCall GJH, Laming DJC, Scott SC
(eds) Geohazards, natural and man-made. Chapman and Hall,
London, pp 5363
Pierson TC (1980) Erosion and deposition of debris flows at
Mount Thomas, North Canterbury, New Zealand. Earth Surf
Process 5 : 227247
Pierson TC (1985) Initiation and flow behavior of the 1980 Pine
Creek and Muddy River lahars, Mount St. Helens, Washington. Geol Soc Am Bull 96 : 10561069
Pierson TC (1995) Flow characteristics of large eruption-triggered
debris flows at snow-clad volcanoes: constraints for debris
flow models. J Volcanol Geotherm Res 66 : 283294
Pierson TC (in press) An empirical method for estimating travel
times for wet volcanic mass flows. Bull Volcanol
Pierson TC, Janda RJ (1994) Volcanic mixed avalanches: a distinct eruption-triggered mass-flow process at snow-clad volcanoes. Geol Soc Am Bull 106 : 13511358
Pierson TC, Scott KM (1985) Downstream dilution of a lahar:
transition from debris flow to hyperconcentrated streamflow.
Water Resour Res 21 : 15111524
Pierson TC, Janda RJ, Thouret JC, Borrero CA (1990) Perturbation and melting of snow and ice by the 13 November 1985
eruption of Nevado del Ruiz, Colombia and consequent mobilization, flow, and deposition of lahars. J Volcanol Geotherm Res 41 : 1766
Pierson TC, Janda RJ, Umbal JV, Daag AS (1992) Immediate
and long-term hazards from lahars and excess sedimentation
in rivers draining Mt. Pinatubo, Philippines. US Geol Surv
Water Resour Invest Rep 924039, 35 pp
Qian Y, Wenhai Y, Wenlin Z, Xiuwen C, Longrong Z, Wengui X
(1980) Basic characteristics of flow with hyperconcentration of
sediment. In: Boning L chairperson) Proc Int Symp River Sedimentation, 2429 March 1980, Beijing, China. Guanghua
Press, Beijing, pp 175184
Rodine JD (1974) Analysis of mobilization of debris flows. PhD
thesis, Stanford Univ, Stanford, 226 pp
Scott KM (1988) Origins, behaviour and sedimentology of lahars
and lahar-runout flows in the Toutle-Cowlitz River system.
US Geol Surv Prof Pap 1447-A: 174
Scott KM, Pringle PT, Valance J (1995) Sedimentology, behavior,
and hazards of debris flows at Mount Rainier, WA. US Geol
Surv Prof Pap 1547 : 156
Scott WE, Iverson RM, Vallance JW, Hildreth W (1995) Volcano
hazards in the Mount Adams Region, Washington. US Geol
Surv Open-File Rep, pp 95492
Sharp RP, Nobles LH (1953) Mudflow of 1941 at Wrightwood,
southern California. Geol Soc Am Bull 64 : 547560
Smyth MA (1991) Movement and emplacement mechanisms of
the Ro Pita volcanic debris avalanche and its role in the evolution of Cotopaxi Volcano. PhD thesis, Univ Aberdeen, Scotland
Trabant DC, Waitt RB, Major JJ (1994) Disruption of Drift glacier and origin of floods during the 19891990 eruptions of
Redoubt Volcano, Alaska. J Volcanol Geotherm Res
62 : 369385
Vallance JW, Scott KM (1997) The Osceola mudflow from Mount
Rainier: sedimentology and hazard implications of a hugh
clay-rich debris flow. Geol Soc Am Bull 109 : 143163
Waitt RB Jr, Pierson TC, MacLeod NS, Janda RJ, Voight B, Holcomb RT (1983) Eruption-triggered avalanche, flood and lahar at Mount St. Helens: effects of winter snowpack. Science
221 : 13941397