Document Actions

IJEGE-11_BS-Loye-et-alii

background image
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
115
DOI: 10.4408/IJEGE.2011-03.B-014
MORPHOSTRUCTURAL ANALYSIS OF AN ALPINE DEBRIS FLOWS
CATCHMENT: IMPLICATION FOR DEBRIS SUPPLY
a. LOYE
(*)
, m. JABOYEDOFF
(*)
, a. PEDRAZZINI
(*)
, J. THEULE
(**)
,
f. LIÉBAULT
(**)
& R. METZGER
(*)
(*)
Institute of Geomatics and Risk Analysis, University of Lausanne, Switzerland
(**)
Cemagref, Unité de Recherche Erosion Torrentielle, Neige et Avalanches, Grenoble, France
Debris flow hazard assessment on DEM
INTRODUCTION
Debris flows transport large amounts of sediment
out of the watershed over geological time scale. The
size of debris/alluvial fans accumulated along valleys
margins attests of the intensity of hillslope processes.
From cursory field observation, erosion processes
leading to debris supply recharging channels seems to
be implicitly linked to the geomorphic and geostruc-
tural conditions prevailing at the source areas (cliffs
and bedrock basin). Recent investigations have dem-
onstrated that bedrock landslides and collapse of cliff
faces by rockfall are a first order control of hillslope
erosion rate (e.g. d
ensmoRe
et alii, 1997; s
CHRodeR
& b
isHoP
, 1998; H
ovius
et alii, 2006). Mass wasting
production depends on one side from the lowering rate
of local base level (e.g. d
avis
, 1899; H
aCk
, 1960): the
valley slopes steepen and become progressively un-
stable and collapse, involving not only the overlying
sediment deposit but also the bedrock (b
uRbank
et
alii, 1996). Rates and mechanism of debris production
are controlled on the other side by the lithologic char-
acteristics of the topography and the geology influenc-
ing the local state of stress (e.g. s
ayCHin
et alii, 1998;
e
Rismann
& a
bele
, 2001; C
Ruden
, 2003). Historical
investigation of small alpine catchment in the Swiss
Alps have shown that periods of high debris flows
magnitude and frequency result from intense rock-
slides events or rock avalanches of Mm
3
(e
isbaCHeR
ABSTRACT
Rock slope instabilities are implicitly linked to
the supply of sediment and debris recharging channels
prone to debris flow. Hence, the incorporation of bed-
rock structure and terrain morphology can be relevant
in the analysis of sediment budget and debris flow
hazard assessment. Here, the mode of debris produc-
tion of the Manival catchment (northern French Alps)
is documented by the study of its morphostructural
aspects extracted from high resolution DEM. Terrain
implication in the process of debris supply is evalu-
ated by: a) A systematic classification of the major
morphological units based on the slope gradient that
enables a spatial analysis of zones of debris produc-
tion and deposition. b) A detailed structural analysis
performed on DEM in order to identify potential un-
stable slopes. c) An analysis of the gullies orienta-
tion that informs in term of structural control of the
sources zones. d) Localisation of high density joints
sets that document about whether sources of continu-
ous debris production are controlled by the structural
setting of the catchment. These DEM-based indicators
can be used as proxies for assessing the influences of
the current topography and enable to quantify a degree
of susceptibility to mass wasting and hillslope erosion
activity. This present contribution suggests some di-
rections for characterizing sediment flux dynamic in
small alpine catchment.
K
ey
words
: Sediment budget, Erosion, Structural control,
background image
A. LOYE, M. JABOYEDOFF, A. PEDRAZZINI, J. THEULE, F. LIéBAULT & R. METZGER
116
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
rocks, the depositional mode of the lithologies (bed-
ding), their tectonic setting (folds, faults, joint sets)
as well as the slope behaviour against mass wasting
and erosion (l
oCat
et alii, 2000). The major mor-
phological units and their morphostructural pattern
are now displayed in details through high resolution
DEM. Thus, they can be identified, which enables to
interpret the role of the major features of topography,
such as the main structural sets shaping the slopes and
the hillslope processes characteristic shapes (e.g. J
a
-
boyedoff
et alii, 2004). The method introduces three
DEM-based indicators that can provide information
on the nature and mode of slope erosion activity.
There are:
- a classified slope steepness map.
- a map of susceptibility index to planar and wedge
failures.
- a mean orientation of zones having high density sets
of joints.
CLASSIFIED MAP OF SLOPE STEEPNESS
The slope steepness influences shear stress in-
tensity upon soil. Hence, the steepness of topog-
raphy is directly involved in slopes stability, soil
creeping and runoff rates. The slope morphology of
a terrain is given by characteristic slope angles. The
s
tRaHleR
’s (1950) law of constancy of slopes states
that slope morphology of a local topography tends to
group prominently around a local mean slope angle
values that are normally distributed with low disper-
sion. Hence, predominant sets of slope angle can be
identified from DEM and related to the major mor-
phological units forming the topography, such as
torrential plain, debris-mantle slope deposits, cliffs
(Fig. 2). Overlying this slope angle classification on
a 3D shaded relief map enables a spatial analysis of
process-related morphology, such as zones of debris
production and deposition.
POTENTIAL ROCk SLOPE INSTABILITIES DE-
LINEATION
The significance of geologic structure on the slopes
stability of bedrock has been demonstrated in the dis-
cipline of rock mechanics (e.g. t
eRzaGHi
, 1962; H
oek
& b
Ray
, 1981; s
elby
, 1982). Studies that have focused
on the structural control of erosion could show the link
between structural geomorphic aspect and mass wasting
(e.g. R
aPP
, 1960; G
eRbeR
& s
CHeideGGeR
, 1973; s
ayCHin
& C
laGue
, 1984; b
aRdou
et alii, 2003). Analyses of
their magnitude have highlighted the coexistence of
two types of processes describing the forming of de-
bris flow volumes (H
aebeRli
et alii, 1991; b
aRdou
&
J
aboyedoff
, 2008):
Continuous debris supply linked to hillslope ero-
sion processes in steady state; denudation behaves
gradually and is related to weathering and rainfall in-
tensity (climate-erosion relationship).
By pulse, through the occurrence of high magni-
tude (extraordinary) debris supply events (e.g. bed-
rock landslides) implying additional geological and
structural predisposition.
Studies focusing on morphometric and topocli-
mate factors to enhance the understanding of the
causal processes governing mass wasting and sedi-
ment flux are often limited by the small scales of in-
vestigation: at local scale, the topography, the tectonic
setting and the climate are rather similar, whereas pro-
duction rates of sediment and mechanism contribut-
ing to accumulate debris can be very heterogeneous
throughout adjacent subcatchments. According to the
fact that rock slope instabilities influence the supply
and volume magnitude of debris, it seems relevant to
incorporate the influence of bedrock structure as well
as terrain morphology into assessment of basin scale
sediment budget and debris flow hazard assessment.
The present study describes a way to document the
mode of debris production by analysing the morpho-
structural aspects of a small alpine catchment prone to
high magnitude debris flows and is illustrated by the
Manival basin, a very active catchment of the Char-
treuse Massif (northern French Alps). The method ba-
ses on topographic elements that can be extracted from
digital elevation model (DEM) and improved with ge-
ological maps. Erosion processes analysis is characte-
rized in terms of process related morphology and slope
failure susceptibility according to the general structural
setting. These can be used as proxies for evaluating ter-
rain implication in the process of debris supply or for
preliminary assessing potential zones of erosion. This
approach suggests a direction for further investigation
on the geomorphostructural role contributing to charac-
terize sediment flux dynamic in small alpine catchment
METHODOLOGY
The topography of a catchment reflects the com-
pounded influence of the mechanical properties of the
background image
MORPHOSTRUCTURAL ANALYSIS OF AN ALPINE DEBRIS FLOWS CATCHMENT: IMPLICATION FOR DEBRIS SUPPLY
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
117
.
IMPLICATION TO DEBRIS SUPPLY
The degree of susceptibility to slope instabilities
can be quantified by its number of detected failure
mechanism per surface area. The density of failure
mechanism can be weighted by an assessment of its
factor of safety. This enables to emphasize potential
unstable zones in comparison to potential stable ones.
When the topography shows a clear structural control
of its morphology, zones of continuous debris supply
can be delineated by assuming that sediment produc-
tion should be located where the density of joints sets
is more important. A susceptibility index of debris
production can be then deduced by combining the
zones of susceptibility to slope failures with the ones
having a high density of joints and limited to a specific
morphological unit.
STUDY AREA
The Manival catchment (Fig. 1 left) is a tributary
to the Grésivaudan Valley located in the eastern bor-
der of the Chartreuse Massif, about 20 km north-east
from the town of Grenoble (France). The Chartreuse
mountain range belongs to the northern subalpine do-
main (or helvetic), that corresponds to the mesosoic
cover of the external alpine crystalline belt. This mas-
sif has formed in an alternation of marls and limestone
ranging from the upper Jurassic to early Cretaceous
age and embedded in layers of decimetric to metric
thickness. It displays a structure of inclined folds from
Miocene age containing important and continuous
overthrust faults in response to the NW-SE crustal
shortening during the Alpine orogene (Gidon, 1991).
& G
aRdneR
, 1983; G
RiffitHs
et alii, 2004). Especially,
the nature of discontinuity sets and their agencement
according to the slope gradient of the topography influ-
ence more the stability of a rock slope than the intrasic
properties of an intact rock mass (e.g. n
oRRisH
& w
illis
,
1996; R
ouilleR
et alii, 1997; H
antz
et alii, 2003). In
the same way, tectonic deformation such as fold position
and relief geometry must be taken into account when
considering rock slope stability (C
oe
& H
aRP
, 2007). In
order to analyse whether this structural geometry con-
tributes to slopes instability and gullies formation across
the catchment, the major discontinuity sets are identified
based on the analysis of the discontinuity surface con-
tained in the topography. Potential unstable slopes and
their mode of failure can be then localized in the catch-
ment by performing a kinematic analysis. This enables
to spatially illustrate where the topography in relation
with the presence of discontinuities is a primary mode
of potential slope mass instability.
STRUCTURAL CONTROL DESCRIPTORS
The influence of the discontinuity pattern on the
current morphology of the catchment can be assessed
using (s
CHeideGGeR
, 1980; b
eavis
, 2000):
The maximum frequency orientation of the gullies.
The orientation of the maximum frequency of dis-
continuities.
These can inform not only in term of structural
control of the cliffs and gullies, but enable to docu-
ment whether sources zones of continuous debris and
screes production are controlled by the structural set-
ting of the catchment
Fig. 1 - (left) Location of study area; the Manival catchment is displayed in light gray showing the impressive debris
fan.(right) Photograph showing the upper Manival basin with the main lithologies and faults sets: J5-6: cal-
careous marl (upper Oxfordian), J7: marly-limestones (Sequanian), J8: stratified limestones interbedded with
marls (kimmeridgian), J9 massif limestones (Thitonian). A is an inverse compressive fault at the top. B and
D are local minor inverse faults.
Φ
M
Manival thrust, Φ
B
Baure thrust. (modified after www.Geol-alp.com (Gidon 1991))
background image
A. LOYE, M. JABOYEDOFF, A. PEDRAZZINI, J. THEULE, F. LIéBAULT & R. METZGER
118
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
SLOPE STEEPNESS ANALYSIS
The slope angles map were created using a ma-
trix of 3x3 DEM cells (b
uRRouGH
& m
C
d
onnell
,
1997) and classified in steps of 1°. The frequency of
occurrence of each class of slope angle was normal-
ized considering their real surface. The slope angle
frequency distribution (SAFD) was decomposed into
several normal distributions, where the sum of those
normal curves fits the SAFD (Fig. 2). This was proc-
essed using the freeware Histofit (l
oye
et alii, 2009),
an Excel
©
-based application that computes the most-
likely normal curves in an iterative way until their
sum fits at best a target function represented here by
a SAFD. The fitting process is done by minimizing
the standard error using optimisation procedures of
the Excel solver. The initial values are defined accord-
ing to the shape of the SAFD, where local maximum
and minimum can be identified visually along the
curvature of the SAFD. The sets of slope steepness
representing a morphological unit are delimited ac-
cording to the slope angle where a normal distribution
becomes dominant over the others. In example in fig.
2, the threshold for the cliffs is over 47°.
Morphostructural analysis and kinematic analysis
The major discontinuity sets of the catchment
were identified based on the discontinuity surface of
the 3D topographic analysis performed with the soft-
ware Coltop 3D (J
aboyedoff
et alii, 2007). Coltop 3D
enables an analysis of the topography by representing
the dip and dip direction of the slopes of a DEM with
a unique colour code, the Hue saturation intensity
system (HIS). The topography is displayed as a 3D
shaded relief and the colour code is ranged in value
of a Schmidt Lambert projection. Considering that
The erosion, especially in the eastern part of the mas-
sif, has since then deeply entrenched the folds, leaving
a dissected relief with steep cliff faces.
The Manival catchment (Fig. 1 right) has devel-
oped in the most southeastern anticline of the massif,
building one of the largest alluvial fan surfaces from
the western Alps made of torrential debris. The valley
covers an area of 3.7 km
2
from the top (1738 m) to the
apex of the alluvial fan. The high ridge and peaks of the
basin are made of massif (Tithonian) to stratified lime-
stones interbedded in its lower parts by layers of marls
(Kimmeridgian). Depending on the marls thickness,
they can form well-defined escarpments. The mid-el-
evation slopes are generally made of marly limestone
from the so-called Sequanian (Oxfordian-Kimmerigdi-
an border). Those layers display another face of cliffs.
The lower part of the catchment is made of calcareous
marls of metric beds, alternating with marls from the
upper Oxfordian. In the down part of the catchment,
they are mostly covered by slope deposits, but exhibit
steep slopes in the upper basin, although they might be
less resistant. The two valley sides correspond to the
limbs of the anticline. The hinge zone is cut by a sys-
tem of two major inverse faults that cut sidelong the
axis of the valley. It results a higher elevation of the
western limb and the east part is reversed, leading to an
inclined fold. Other minor faults exist. All those faults,
as the lithologies, have probably played a major role
in the impressive headward entrenchment depth of the
catchment. Today, the erosion is still very active, but
concentrated essentially in the upper catchment, at the
bedrock basin displaying steep cliff walls. Steep slope
gradient coupled with a very low vegetation cover and
low permeability induce an important runoff as a re-
sponse to heavy rainstorms that occur regularly up there
due to enhanced convective cells. This is also responsi-
ble for an important bedload transport, erosion of screes
covered slopes along the channel, interfluve hillslope
transport of sediment and initiation of debris flows. The
middle and down part of the watershed, although quite
steep as well, is well vegetalized. Their contribution in
term of erosion activity must be therefore reduced.
DATA PROCESSING
The study was conducted with a 1m cell size
DEM derived from airborne LiDAR survey of the en-
tire catchment operated in 2009.
Fig. 2 - Slope angle frequency distribution (SAFD) of
the Manival catchment and decomposition in
major morphological units (displayed in fig.
3); 47° is a threshold above which the slopes
can be considered as cliffs faces
background image
MORPHOSTRUCTURAL ANALYSIS OF AN ALPINE DEBRIS FLOWS CATCHMENT: IMPLICATION FOR DEBRIS SUPPLY
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
119
m
2
. However, only flow paths lying above the torren-
tial plain were kept for the analysis, as below they are
not structurally controlled. The frequency of the gul-
lies geometry was plotted in a stereonet correspond-
ing to a histogram in 3D and the density stereonet of
the discontinuity sets were computed with the H
udson
& P
Riest
s
(1983) method considering a similar spac-
ing for all sets. The distribution of the discontinuity
density was classified in two classes, the threshold be-
ing the density value reaching 95% of the cumulative
distribution. The weighted density of potential slope
failures was defined by dividing the average number
of intersections per DEM cell by its factor of safety
based on the limit equilibrium analysis (tab. 1), as-
suming that sliding is resisted only by friction similar
for all planes. In this study case, the weighted relative
density index for both planar and wedge are classified
in three sets were their cumulative distribution defined
at the 68-quantile, respectively 95-quantile.
RESULTS AND ANALYSIS
SLOPE ANGLE FREQUENCY DISTRIBUTION
ANALYSIS
The decomposition of the SAFD with the tool
Histofit was achieved using 5 normal distributions (Fig.
2). Several initial values were tested, which yield to
similar normal curves pattern for most of the cases. The
sum of the normal curves reproduces the SAFD of the
Manival topography with a coefficient of determination
DEMs display surface features that can reflect several
structural information, Coltop 3D enables a detailed
investigation of the discontinuity sets shaping the rock
slope of a catchment. When such analysis is based on
airborne DEM acquisition, slopes with overhanging
structures are not detected. They can however be iden-
tified in some other parts of the slope where they out-
crop. The structural setting was then compared with
field observation and terrestrial laser scanner data.
The identified discontinuity sets were compared
with the entire topography using the software Mat-
terocking (J
aboyedoff
et alii, 2004). This GIS tool
enables to assess potential planar and wedge sliding
area on DEM and gives the average number of fail-
ure mechanism per DEM cell, according to a given
spacing between discontinuities. All discontinuities
were processed according to their mean azimuth and
dip. An arbitrary spacing of 1 m (corresponding to the
DEM resolution) was assumed for all discontinuity
sets. Wedges with plunge axes <20° were considered
as marginally potential and not further considered.
STRUCTURAL CONTROL ASSESSMENT
The distribution of the gullies geometries was ex-
tracted from the virtual rivers of the DEM. Flow paths
were determined by the D8 algorithm (o’C
allaGHan
et alii, 1984) considering a contributing area of 5000
Tab. 1 - Factor of safety (FS) computation (N
orriSh
&
w
illie
, 1996
Tab. 2 - Threshold slope angles defining Morphologi-
cal units
Fig. 3 - Classified slope steepness map highlighting the
major morphological units. Note the rugged
torrential plain, the debris cones on the east
side and the slope bottom deposit on the west
side of the valley
background image
A. LOYE, M. JABOYEDOFF, A. PEDRAZZINI, J. THEULE, F. LIéBAULT & R. METZGER
120
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
close to 1, meaning that the morphology is clearly con-
strained. Threshold slope angles delimiting the sets of
slopes steepness are summarized in table 2. Overlying
this classified slope steepness map with the shaded relief
map (Fig. 3) enables to identify 4 morphological units:
The lowest set corresponds to the torrential plain.
This range of slope steepness delineates the torrential
plain full of sediment deposited throughout the valley
bottom.
The 2
nd
set of steepness displays the debris cones
deposited at the mouth of the tributary gullies and the
bottom-slope deposit cover located at the margin of
the valley bottom.
The 3
rd
range of steepness covers all talus slope
deposits and outcropping bedrock slopes lightly cov-
ered with vegetation. They are located both downs-
lope and upslope from cliff faces.
The steepest sets delineate the cliffs, rocky es-
carpment and the steep bedrock slopes. Two sets of
cliffs were input for a better fitting of the SAFD.
MORPHOSTRUCTURAL ANALYSIS
The Coltop 3D analysis has identified 7 major
topographic orientations (Fig. 4) spatially well dis-
tributed and continuously present in the Manival
catchment (Fig. 5). Four discontinuity sets (J5, J6, J7
and J1) are mainly encountered in the west side of the
valley, whereas J1 is a well-defined discontinuity en-
countered on both sides to the top of the catchment. J5
possesses a more dispersed orientation and develops
exclusively in the deep vertical gullies of the western
slopes. It is possible that J5 represents only a steep
geomorphic features influenced by gully activity, but
may not belongs to J6, for example, which is very
constant with its orientation ranging of ±14 around its
mean value. The east side of the catchment are shaped
by J1 and J4. J7 on the west and J2 and J3 on the east
part define rather clear geomorphic features as they
displays most of the cliffs facing the valley bottom of
the entire catchment. The bedding is gently dipping in
the slope (anaclinal) in most of the catchment.
Bedding field measurements give out a mean
orientation around 270°/35° for the west slopes and
around 100°/45° for the east slope down the valley but
becomes much more variable in the upper catchment
and around the local zones of deformation. But as the
bedding planes could not be well identified with COL-
TOP 3D, they were not taken into account in this study.
They are however important in the west slopes as they
can induce toppling (not yet processed in this study).
The structural scheme is illustrated on pictures (Fig. 6).
Fig. 5 - Discontinuity sets extracted from DEM using
COLTOP 3D. The geometry of each set is given
with its variability of 2σ (see Tab. 3)
Fig. 4 - Discontinuity sets extracted from DEM using
COLTOP 3D. The black circle represents the
variability (2σ).Values are given in Tab 3
Tab. 3 - Discontinuity sets for the Manival catchment
background image
MORPHOSTRUCTURAL ANALYSIS OF AN ALPINE DEBRIS FLOWS CATCHMENT: IMPLICATION FOR DEBRIS SUPPLY
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
121
STRUCTURAL CONTROL DESCRIPTORS IN-
TERPRETATION
The frequency stereonet of the gullies orientations
(Fig. 8) gives two major poles. The pick 107°/44° cor-
responds to the mean dip/dip direction of the west side
catchment flow paths. Comparing this pick value with
the geometry of the discontinuities identified on these
slopes shows that the gullies follow rather the slope
gradient than one preferential discontinuity configura-
tion. In the contrary, the pick value 248°/45° is linked
to the east side and suggests that gullies are strongly
influenced by J4 and its wedges formed with J3 (and
J2 accessory). This value is slightly different from the
greatest slope gradient, implying a potential structural
control of the flow paths.
The maximum density of discontinuities assum-
ing an identical mean spacing for all joints sets leads
to an orientation of about 200°/70° with certain vari-
ability of 20° (Fig. 9). Considering that J1 and J6 are
very persistent over the all massif and that J3^J4 con-
trols the topography of the east side, the mean spac-
ing of these three discontinuities were doubled. The
POTENTIAL FAILURE MECHANISM AND LO-
CALISATION
The map of potential planar sliding shows a high
susceptibility for the steep slopes facing southeast,
which are concentrated mainly on the top central
part (Fig. 7 left). These imply joints set J1, J2 and J3
predominantly. On the west side of the valley, gullies
flanks facing north are exposed to planar failures. The
slope stability seems to be controlled by J6 in this
area. On the east side, they concentrate on the steep
slopes only. Zones of potential wedges are widespread
in the catchment according to the joints sets of this
study (Fig. 7 right). However, its weighted relative
density is very low for most of the catchment. High
susceptibility to wedge sliding is encountered mainly
in the steep slopes facing S-SE, but not only. Several
steep enough slopes, like the ones surrounding the
major gullies on the west side display a relatively mid-
dle high susceptibility. They are controlled by wedges
set configuration J5^J6, J5^J7 and J6^J7
Fig. 6 - Field illustration of the structural scheme
detected on DEM. The location of the photo-
graph is displayed in fig. 5
Fig. 7 - Classes of suscep-
tibility to planar
(left) and (right)
wedge sliding ac-
cording to their
weighted relative
density
Fig. 8 - Frequency stereonet of gullies orientation
(picks red cross) and comparison with the sets
of discontinuities
background image
A. LOYE, M. JABOYEDOFF, A. PEDRAZZINI, J. THEULE, F. LIéBAULT & R. METZGER
122
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
orientation of the maximum density is then situated
around 205°/64°, relatively closed to the previous
value. Zones of high density joints sets are oriented
S-SW and follow most of the cliffs surrounding tribu-
tary gullies on both sides of the valley (Fig. 10).
DISCUSSION
The morphological units extracted from the
SAFD analysis provides a systematic approach for
delineating debris sources areas such as the cliffs and
zones of deposition and remobilisation such as slopes
deposit and alluvial plain. The threshold slope angle
above which the slopes can be considered as cliff fac-
es and outcropping bedrock is 47°. This is comparable
to what has been obtained in the Helvetic and Ultra-
helvetic Swiss Alps (l
oye
et alii, 2009). Combining
this slope angles classification with field observation
of the morphology can inform about the mode of de-
bris production and transportation. The surface of the
torrential plain is rugged (Fig. 3), showing several
remnant channel beds. This high amount of old chan-
nels can attest of high magnitude debris flow events in
order to change the course of the main channel. The
mean steepness of the torrential plain is 11° (20%),
which corresponds to the mean angle of depose of
granular debris flow type (R
iCkenmann
, 1995). The
bottom-slope deposit morphological unit can be sepa-
rated in two types according to their form and their
clear break of slope angle at the interface with the tor-
rential plain (Fig. 3):
1. Cone-shaped deposits at the mouth of ephemeral
gullies; their relative steep slope of deposition
suggests a rockfall-dominant type of deposit.
Water-dominated flow and debris flow fans are
usually less steep, because they are composed of
less thicker and coarser material. These cones are
entrenched by ephemeral channels.
2. Sediment deposit at the toe of the talus slopes, gen-
tly flatten footslopes made of rock fragments and
covered with mature vegetation suggest progres-
sive deposition of mode wash slope.
The upper catchment displays no cone-shaped or
bottom-slope deposit at the toe of gullies and cliffs
showing an absence of long term mass wasting de-
posit coming from the sides of the valley and high
connectivity of the entire upper catchment to transport
debris away. Debris flow frequency seems therefore to
be supply-limited.
The 3D analysis of the major topographic orien-
tation performed with Coltop 3D enables to identify
the major discontinuities structuring the topography
Fig. 9 - Geometry of the maximum frequency of the dis-
continuity pattern of manival catchment
Fig. 10 - Map of the zones having a high density sets of
joints
Fig. 11 - Cumulative distribution of the weighted density
of potential wedge sliding intersection (shown
in fig. 7) and the fitted Poisson distribution
background image
MORPHOSTRUCTURAL ANALYSIS OF AN ALPINE DEBRIS FLOWS CATCHMENT: IMPLICATION FOR DEBRIS SUPPLY
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
123
ness may however bias the analysis, because the gul-
lies are clearly chopped by J1 and J6.
The analysis of the maximum density of disconti-
nuity reveals mean orientations defining steep slopes
surfaces that are facing S-SW. They follow alongside
the gullies particularly well and on both side of the
catchment. Hence, production of continuous rockfall
screes and debris must be located where the zones
with high density of discontinuities are.
The susceptibility index of debris production is
obtained by multiplying this apparent density of dis-
continuity with the weighted relative density of fail-
ure mechanism (sum of potential planar and wedge
failures) (Fig. 12). This is classified in three sets
(low, mid and high susceptibility) where the cumula-
tive distribution defines the 50-quantile, respectively
75-quantile, as suggested by comparing the index with
orthophotos and field observation (Fig. 13).
.
CONCLUSION AND OUTLOOK
Observation and investigation on shaded high
resolution DEM of the major morphological units
delineated with the classified slope angle map en-
able to identify evidences of debris production and
deposition such as rockfall deposit and debris flow
activity. Sediment slopes accumulation in contact
with channels and gullies can inform on the sediment
supply dynamic. The DEM-based indicators devel-
of the catchment. Representing the pattern of dis-
continuities on the shaded relief map shows that they
cover about 25% of the computed horizontal surface
and are spatially well distributed, meaning that the
topography of the catchment is rather shaped by this
structural setting. J1 and J6 are clearly involved in the
shape of large tributary gullies. J1 seems to be very
persistent, as it is found on both side of the valley. J4
seems to control large slope surfaces of the east side
of the valley and is cut laterally by J1 and J2 to form
a succession (en echelon) of small gullies and v-shape
geomorphic features. J3 and J7 display clear and con-
tinuous geomorphic features as they draw most of the
cliffs facing the valley bottom. In the top central part,
J1 and J7 control the rock walls, whereas J4 and J6
clearly control the morphology in a wedge like con-
figuration and confirm that they are true discontinuity
sets. Photograph analysis show that J4 represents here
the bedding plane but this is not the case anymore as it
goes away from the top central part and on the side of
the valley. The weighted relative density of joints that
fulfils the conditions for potential planar and wedge
failure show that the central top part of the catchment
is well exposed to slope instabilities. They are pro-
duced on joint sets J1, J2 and J3. Potential planar slid-
ing zones follow the south flanks of the large gullies
situated on the western part and can be a mode of de-
bris supply into the channels. High weighted relative
density of wedges is located principally (but not only)
in the slope facing S-SE as is it the case for planar
sliding. Cliffs surrounding gullies with such an orien-
tation are susceptible to experiment high debris sup-
ply. In this study, the weighted relative density of both
failure mechanisms could be modelled by a Poisson
distribution (Fig. 11). The widespread potential wedge
sliding demonstrates that the present discontinuities
sets play an important role in the slope stability of
the catchment, even though most of the zones have a
low susceptibility. This can be partly explained by the
wedge-like configuration of the east and central part
of the topography mentioned above. The influence of
discontinuities, such as faults and fractures, can be
underlined by the frequency stereoplot of the gullies
geometry. The gullies of the upper eastern catchment
part are directed principally in the same direction of
J4 and oriented within the wedges caused by J3^J4,
showing a clear structural control. The western side is
apparently not structurally controlled. Its great steep-
Fig. 12 - Map of susceptibility to debris production ac-
cording to the combination of the three DEM-
indicators processed in this study (yellow=low,
blue = mid, red = high susceptibility)
background image
A. LOYE, M. JABOYEDOFF, A. PEDRAZZINI, J. THEULE, F. LIéBAULT & R. METZGER
124
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
oped in this study can assess the influence of the
morphostructural aspect on the current morphology
of the catchment and enable to quantify a degree of
susceptibility to mass wasting and hillslope erosion
activity. Processes related to rock slope instability
and continuous debris production are potentially dis-
tinguishable by assuming that continuous sediment
supply must be located where high density joints
sets are. All these affirmations must be verified with
field measurements. The erosion activity of cliffs and
gullies of the upper rock basin at Manival has been
monitored periodically with terrestrial laser scanner
since 2009 and should provide volume and magni-
tude of sediment to quantitatively highlight the main
geomorphic processes controlling the production
of debris. So far, this approach was compared with
aerial photo analysis and field observation only (Fig.
13), but first results are promising.
ACKNOWLEDGEMENTS
The first author would like to address his grati-
tude to M.-H. Derron and B. Matasci from the UNI
Lausanne for their constructive comments and ideas
during the development of this study.
Fig. 13 - Two erosion scarps arising recently and supplying the channel with debris as field observation that enable to
evaluate the DEM-indicators. (leftt) Close-up of fig. 12 displaying the susceptibility index to debris production
on orthophoto(© IGN). (right) Field photographs taken in summer 2010
REFERENCES
b
aRdou
e., f
ouRnieR
f. & s
aRtoRi
m. (2003) – Paleofloods reconstruction on Illgraben torrent (Switzerland): a need for today
frequency estimation. Intern. Workshop on Paleofloods, Historical Data & Climate Variability: Application in Flood Risk
Assessment, Barcelona.
b
aRdou
e. & J
aboyedoff
m. (2008) – Debris flows as a factor of hillslope evolution controlled by a continuous or a pulse proc-
ess? From: G
allaGHeR
k., J
ones
s. J. & w
ainwRiGHt
J. (eds) - Landscape evolution: Denudation, Climate and Tectonic
Over Different Time and Space Scales. Geological Society, London, Special Publication, 296: 63-78.
b
eavis
S. G. (2000) - Structural controls on the orientation of erosion gullies in mid-western New South wales, Australia. Geo-
morphology, 33: 59-72.
b
uRbank
d.w., l
eland
J., f
ieldinG
e., a
ndeRson
, R.s. b
RozoviC
, n., R
eid
m.R. & d
unCan
, C. (1996) - Bedrock incision, rock
uplift and threshold hillslopes in the northwestern Himalayas. Nature, 379: 505–510.
b
uRRouGH
P. & m
C
d
onnell
R. a. (1997) - Principles of Geographical Information Systems. Oxford University, 330 pp.
C
oe
J.a. & H
aRP
e.l. (2007) – Influence of tectonic fording on rockfall susceptibility. American Fork Canyon, Utah, USA. Nat-
Hazards Earth Syst. Sci., 7: 1-14.
C
Ruden
d.m. (2003) - The shapes of cold, high mountains in sedimentary rocks. Geomorphology, 55: 249-261.
d
avis
w.m. (1899) - The geographical cycle. Geography Journal, 14: 481-504.
d
ensmoRe
a.l., a
ndeRson
R.s., m
C
a
doo
b.G.
and
. e
llis
m.a (1997) - Hillslope evolution by bedrock landslides. Science, 275,
369-372.
e
isbaCHeR
G.H. & C
laGue
J.J. (1984) – Destructive mass movement in high mountains: hazard & management. Geol. Survey
of Canada, 84: 16.
e
Rismann
t. H. & a
bele
G. (2001) - Dynamics of rockslides and rockfalls. Springer, Berlin, 316, 2001.
G
eRbeR
e. & s
CHeideGGeR
a.e. (1973) - Erosion and Stress-induced features of steep slopes. Z. Geomorphol, N.F., Suppl. Band
18: 38-49.
background image
MORPHOSTRUCTURAL ANALYSIS OF AN ALPINE DEBRIS FLOWS CATCHMENT: IMPLICATION FOR DEBRIS SUPPLY
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
125
G
idon
m. (1991) - Géologie de la Chartreuse - Sentiers de la Chartreuse. Circuit de la Dent de Crolles. Association “A la décou-
verte du Patrimoine de Chartreuse”, publ. 1d, 1° éd., 20 p., 9 fig., www. Geol-alp.com.
G
RiffitHs
P.G. & w
ebb
R.H. (2004) - Frequency and initiation of debris flows in Grand Canyon, Arizona. J. Geophys. Res.,
109, F04002.
H
aCk
J. t. (1960) - Interpretation of erosional topography in humid temperate regions. American Journal of Science, 258-A:
80-97.
H
aebeRli
w., R
iCkenmann
d. & R
össli
u. (1991) – Murgänge, Ursacheanalyse der Hochwasser 1987. Ergebnisse der Untersu-
chung. Mitt. No. 14. Landeshydrologie und –geologie, Bern.
H
antz
d., v
enGeon
J. m. & d
ussauGe
-P
eisseR
C. (2003) - An historical, geomechanical and probabilistic appoach to rockfall
hazard assessment. NHESS, 3: 693-701.
H
oek
e. & b
Ray
J.w. (1981) - Rock Slope Engineering. The Institution of Mining and Metallurgy, London. 525 pp.
H
ovius
n. & s
taRk
C.P. (2006) - Landslide-driven erosion and topographic evolution of active mountain belts. In: e
vans
S.G. et
alii (Eds.). Landslides from massive rock slope failure: 573-590, Springer.
H
udson
J.a. & P
Riest
s.d. (1983) - Discontinuity frequency in rock masses. International Journal of Rock Mechanics and Mining
Sciences & Geomechanics Abstracts, 20: 73–89.
J
aboyedoff
m. (2002) - Matterocking v 2.0: A program for detecting rockslide instabilities (available on www.crealp.ch).
J
aboyedoff
m., b
aillifaRd
f., P
HiliPPossian
f. & R
ouilleR
J.-d. (2004) - Assessing fracture occurrence using “weighted fractu-
ring density”: a step towards estimatin rock instability hazard. NHESS, 4: 83-93.
J
aboyedoff
m., m
etzGeR
R., o
PPikofeR
t., C
outuRe
R., d
eRRon
m.-H., l
oCat
J. & t
uRmel
d. (2007) - New insight techniques to
analyze rock-slope relief using DEM and 3D-imaging cloud points: COLTOP-3D software. In: e
beRHaRdt
e., s
tead
d &
m
oRRison
t.: Rock mechanics (2007, eds) - Meeting Society’s Challenges and demands (Vol. 2), Taylor & Francis: 61-68.
l
oCat
J., l
eRoueil
s. & P
iCaRelli
l. (2000) - Some considerations on the role of geological history on slope stability and estima-
tion of minimum apparent cohesion of rock mass. In: b
RomHead
e., d
ixon
n. & i
bsen
m.l. (eds) - Landslides in research,
theory and practice. 8
th
International Symposium on Landslides in Cardiff: 935-942.
l
oye
a., J
aboyedoff
m. & P
edRazzini
a. (2009) - Identification of potential rockfall source areas at a regional scale using a
DEM-based geomorphometric analysis. Nat. Hazards Earth Syst. Sci., 9: 1643-1653.
n
oRRisH
n.i. & w
yllie
d.C. (1996) - Rock slope stability analysis. Transportation Research Voard. Special Report, Serial. Pub-
lished in Washington D. C., 247, 509-513.
o’C
allaGHan
J.f. & m
aRk
d.m. (1984) - The extraction of drainage networks from digital elevation data. Computer Vision,
Graphics, and Image Processing 28: 323-344.
R
aPP
a. (1960) - Talus slopes and Mountain walls at Tempelfjorden, Spitsbergen. Nor. Polarins. Skr. 119, 96 pp.
R
iCkenmann
d. (1995) - Beurteilung von Murgängen. Schweizer Ingenieur und Architekt, 48: 1004-1008.
R
ouilleR
J.-d., J
aboyedoff
m., m
aRRo
C. & P
HiliPPossian
, f. (1997) - Matterock: méthodologie d’étude d’instabilités de falaise
et d’appréciation du danger. Publications de la Soc. Suisse de Méc. Sols et roches, 135: 13-16.
s
auCHyn
d. J. & G
aRdneR
J. s. (1983) - Morphometry of open rock basins, kananaskis area. Canadian Rocky Mountians. Can.
J. Earth Sci., 20: 409-419.
s
auCHyn
d.J., C
Ruden
d.m. & H
u
x.Q. (1998) – Structural control of the morphometry of open rock basins, kananaski region,
Canadian Rocky Mountains. Geomorphology, 22: 313-324.
s
CHeideGGeR
e. a. (1980) - Alpine joints and valleys in the Neotectonic stress-field. Rock Mech. Suppl., 9: 109-124.
s
HRodeR
J. f. & b
isHoP
m. P. (1998) - Mass movement in the Himalaya: new insights and research directions. Geomorphology
26: 13-35.
s
tRaHleR
a.n. (1950) - Equilibrium theory of erosional slopes approached by frequency distribution analysis. American Journal
of Science, 248: 673-696, 800-814.
s
elby
m.J. (1982) - Controls on the stability and inclinations of hillslopes formed on hard rock. Earth surface Processes and
Landforms: 449-467.
t
eRzaGHi
k. (1962) - Stability of slopes on hard unweathered rock. Geotechnique, 12: 251-263.
Statistics