Document Actions

IJEGE-11_BS-Fiorillo&Guadagno

background image
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
53
DOI: 10.4408/IJEGE.2011-03.B-007
RAINFALL INITIATION OF DEBRIS FLOWS IN CAMPANIA (ITALY):
A TWO-PHASE ANALYSIS
f
RanCesCo
FIORILLO & f
RanCesCo
M. GUADAGNO
(*)
(*)
Dipartimento di Studi Geologici e Ambientali, Università degli Studi del Sannio, via dei Mulini 59/9, 82100 Benevento, Italy
E-mail: francesco.fiorillo@unisannio.it; guadagno@unisannio.it.
rain storms, triggering hundreds of initial instabilities,
such as debris slides, some of which then evolve into
debris avalanches and debris flows, causing devasta-
tion in the lower zones.
The most catastrophic events were those of Octo-
ber 1954 along the Salerno coast (P
enta
et alii, 1954)
and that of May 1998 (d
el
P
Rete
et alii, 1998; d
e
R
iso
et. alii, 1999; G
uadaGno
& P
eRRiello
z
amPelli
, 2000;
b
RanCaCCio
et alii, 2000; P
aResCHi
et alii, 2000; R
evel
-
lino
et alii, 2004; G
uadaGno
et alii, 2005) that struck
Sarno, Quindici, Siano and Bracigliano. Other cata-
strophic events occurred along the Sorrento peninsula
in February 1963 (m
ele
& d
el
P
Rete
, 1999), March
1969 (C
ivita
et alii, 1975), January 1997 (C
alCateRRa
& s
anto
, 2004). In December 1999 catastrophic debris
flows occurred at Cervinara, along the northern ridge
of Mt. Partenio (f
ioRillo
et. alii, 2001), and in March
2005 debris flows occurred at Nocera.
f
ioRillo
& w
ilson
(2004) analysed rainfall data
from the (Italian) National Hydrographic Service
from the principal storms that triggered debris flows
in Campania area up to 1998, supplemented by un-
published hourly rainfall data from specific storms.
Rainfall data have been analyzed to understand
the hydrological process that leads to debris flow
initiation, and two main stages were proposed about
rainfall accumulation in the soil:
- Early wet season recharge of unsaturated soil
moisture, reducing the moisture deficit (soil suc-
tion) induced by the long dry season, without de-
ABSTRACT
Debris flows in Campania begin as soil slides
within the pyroclastic mantle that blankets the steep
local hill slopes, which are, in turn, composed largely
of carbonate bedrock. The historical pattern appears to
be that clusters of debris avalanche-flows occur after
intense rainstorms that follow an accumulation of a
certain amount of pre-storm seasonal rainfall.
Soil moisture appears to show a seasonal pattern
of summer drying and winter wetting that is typical
for the Mediterranean climate of the region. The phys-
ical analysis of the interaction of rainfall with the py-
roclastic mantle requires two-phase approach. Phase
1, early in the rainfall season, concerns the accumu-
lation of the retention water in the soils, up to field
capacity, whereas the second phase of the analysis
examines the accumulation of surplus moisture from
intense rainfall, leading to the development of positive
pore pressures and debris flow initiation. Rainfall data
analysed cover also the last decade characterised by
no debris-flow activity in the Sarno area
K
ey
words
: rainfall, debris flow, Campania
INTRODUCTION
Debris avalanche-flows occurred many times in
western Campania, involving thin pyroclastic layer
mantling steep limestone slopes, some steeper than
40°. These deadly events occur generally when pe-
riods of prolonged rainfall are followed by intense
background image
FRANCESCO FIORILLO & FRANCESCO M. GUADAGNO
54
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
SOME ASPECTS OF MATERIALS, DE-
BRIS FLOWS AND STORM EVENTS
Complex system of highlands surrounding the
Campania Plain characterises the western sector of
Campania, and along the coast volcanic structures of
Vesuvius and the Phlegrean Fields are present (Fig. 1).
The highlands are formed predominantly by lime-
stone sequences (Mesozoic age), bounded by faults of
Quaternary tectonic activity, forming high ridges and
steep slopes (b
RanCaCCio
et alii, 1978).
Volcanic activity of Vesuvius and the Phlegraean
Fields (R
osi
& s
bRana
, 1987; s
antaCRoCe
, 1987) has
produced air-fall tephras, deposited in function of the
wind direction and accumulated progressively onto
the steep limestone slopes. As a consequence of inter-
mittent eruptions and pedological alteration between
deposition events, the pyroclastic mantle is generally
composed of several irregular, ashy pumiceous layers,
alternating with buried soil
e
sPosito
& G
uadaGno
(1998) stressed the pecu-
liar physical characteristics of the pumices, character-
ised by interconnected internal voids controlling wa-
ter accumulation. Besides, being thixotropic, all soils
have high water retention values and low values of dry
bulk density (γ
d
); then the weight of retained water can
be higher than dry soil weight (t
eRRibile
et alii, 2000).
Figure 2 provide a schematic profile of these
slopes, and table 1 shows some geotechnical charac-
teristics of the pyroclastic mantle.
Some instabilities develop along gullies, due to
erosion of the colluvial deposits (reworked pyroclas-
tic), while a large proportion originate as translational
velopment of positive pore pressure.
- Storm rainfall retained in open pore spaces later in
the wet season, after the soil moisture saturation is
restored, with consequent development of positive
pore pressures.
In this study we analyse hydrological data up
to December 2008, providing further 10 years of
records, which allow us to verify a previous hy-
pothesis on the soil moisture accumulation into the
soil. In particular, we focused the analyses in the
Sarno area (P.zzo d’Alvano), where a high-elevation
rain-gauge is available since May 1998, which al-
lows also to compare data with low-elevation rain
gauges. The topic assumes particular significance
in this case, because the storm of May 1998 didn’t
show exceptional rainfall at any of the rain gauge
stations, to justify the high number of debris flows
produced.
Besides, some details will be further given about
the process of water accumulation into the soil which
can lead to instability, controlled by the evapotranspi-
ration processes and daily rainfall distribution.
Fig. 1 - (a) Italian peninsula and area location; (b) Geo-
logical sketch map of the western side of the
Campanian Apennines (modified from Fiorillo &
wilson, 2004). 1) Alluvial and other continental
deposits (Plio-Pleistocene); 2) Marine terrig-
enous sequences (Mio-Pliocene); 3) Carbonate
unit (Trias-Paleocene); 4) Volcanic area (Quater-
nary); Rain gauge network is shown by triangles
Tab. 1 - Porosity and some geotechnical characteristics of
the pyroclastic soil. Bw, weathered and remoulded
ash deposits; C, pumice level; Bt, weathered ash
level; n, total porosity; neff, effective porosity (ex-
tracted from f
iorillo
& w
ilSoN
, 2004)
Fig. 2- Schematic profile of a typical slope of high-elevated
zones of the Sarno area (overlapping and thick-
ness of the levels are schematised)
background image
RAINFALL INITIATION OF DEBRIS FLOWS IN CAMPANIA (ITALY): A TWO-PHASE ANALYSIS
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
55
flow storms (Fig. 3) have been produced by Kriging
interpolation. High-elevation rain gauges include those
located (Fig. 1) west of Avellino (M.Vergine, 1287 m
a.s.l.), at Vesuvius (Osservatorio, 610 m a.s.l.) and on
the Sorrento Peninsula (Agerola, 691 m a.s.l.; Chiunzi,
617 m a.s.l.). The role of the topography appears very
important in controlling the distribution of the rainfall
of the 1954 storm, but it seems minor for the 1998
storm. In any case, the rain gauge network describes
the spatial rainfall distribution of the rainfall.
The 26 October 1954 occurred after a long dry
summer, and was characterised by a very concentrated
cell located on the Salerno town. No rainfall (or just
few millimetres) was recorded several kilometres in the
northern (Vesuvius area) and southern (Sele plain) sec-
tor, highlighting the “tropical” character of this storm.
The maximum was recorded at Salerno rain gauge with
504 mm (locally the annual mean rainfall is 1250 mm).
The storm had a duration of about 16 hours; however,
more than 75% of the total amount occurred in only 7
hours (data in SIMN, 1954).
The May 1998 storm was less intense than other
storms (Tab. 2), but caused numerous large-scale de-
bris flows. b
RaCa
& o
noRati
(1998) described the sin-
gular characteristics of this storm, such as the unusual
presence of heavy and sustained precipitation at the
beginning of the month of May (9 rainy days). They
pointed out that this heavy spring rainfall was the his-
slides of pyroclastic mantle as complex landslides
(C
Ruden
& v
aRnes
, 1996), from debris slide, to debris
avalanche, to debris flow (f
ioRillo
et alii, 2001). Fol-
lowing classification by H
unGR
et alii (2001), these
landslides may be classified as debris avalanches and
debris flows.
The initial movement occurs mainly above and
below discontinuities along the slopes, such as track-
ways (d
el
P
Rete
et alii, 1998; G
uadaGno
& P
eRRiello
z
amPelli
, 2000; f
ioRillo
et alii, 2001; G
uadaGno
et
alii, 2003-2005), and the sliding surfaces were located
prevalently within a pumice level (d
el
P
Rete
et alii,
1998; C
alCateRRa
et alii, 1999; f
ioRillo
et alii, 2001;
d
el
P
Rete
& d
el
P
Rete
, 2002; G
uadaGno
et alii, 2005;
C
Rosta
& d
al
n
eGRo
, 2003). Following detachment,
the landslide progressively involves the pyroclastic
mantle of the slope, transforming the landslide into
a debris avalanche, with increasing velocity and vol-
ume. At the base of the slopes, the material can reach
long distances (R
evellino
et alii, 2004), and causes the
major damages. Table 2 lists the principal debris flow
events that have occurred in Campania since 1954
Some of the storms that led to numerous debris
flows involved a wide area (up to 100 km
2
) that in-
cluded at least one rain gauge. In these cases, the rain
gauge network was able to record the heaviest rainfall.
In other cases, such as single debris-flow events, the
amount of rainfall associated with the debris flows ac-
tivity is more difficult to ascertain.
Isohyets for the two main rainfall induced debris
Tab 2 - Debris flow events occurred in Campania since
1954
Fig 3 - Isohyets of the storms inducing the main debris
flow phenomena
background image
FRANCESCO FIORILLO & FRANCESCO M. GUADAGNO
56
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
The first rainfall at the beginning of the wet sea-
son occurs in conditions of minimum soil water content
and consequently, a high soil moisture deficit. Thus,
early seasonal rainfall is mainly absorbed into the soil,
increasing the soil moisture. Some intense historical
storms, extracted from the historical series, failed to in-
duce debris flows. In particular, heavy storms occurred
in January 1973 and November 1985 failed to trigger
debris flows, although they induced floods and erosion
in many places. The same amount of rainfall occurring
later in the wet season, when the soil moisture
content is closer to saturation, would interact with
the soil in a much different way, with the rain water
now free to flow down into deeper layers or to move
downslope as runoff or throughflow.
Until field capacity is reached, the water retained
within the soil layer reflects the balance between cu-
mulative rainfall and water loss by evapotranspiration.
Once this moisture level is reached, however, any ad-
ditional infiltration from rainfall is subject also to drain-
age, either directly into the underlying bedrock (perco-
lation) or laterally along the hillslope (throughflow).
Positive pore pressures may be created if the rate of
rainfall infiltration exceeds the rate of drainage, either
because of intense rainfall or reduced permeability in
an underlying "perching" layer.
To restate the above in a different way, the water
content of the soil may be divided into two different
components
PHASE I: ANTECEDENT SOIL MOISTURE
f
ioRillo
& w
ilson
(2004) fixed the soil under-
going evapotranspiration at h=100 cm, because plant
torical maximum for this time of year. Other authors
suggested that the rain gauge network might have
been insufficient in defining the characteristics of the
storm (m
azzaRella
et.alii, 2000).
After the debris flow event of May 1998, a high-
elevated rain gauge was located on the upland of Pi-
ani di Prato (849 m a.s.l.), above the detachment of
the landslides of May 1998. Besides, between 1971
and 1977 a high-elevation rain gauge was in opera-
tion (Pozzo S.Romano, 752 m s.l.m), sited along the
water divide between Lauro and Sarno villages. Fig-
ure 4 shows the scatter plot of 2-days rainfall of the
low-elevated rain gauges (Quindici and Lauro) and
the high-elevated rain gauge (Piani di Prato and Poz-
zo S.Romano). The scatter plot indicates that data not
dispersed, and provide two main considerations on the
rainfall spatial distribution:
- rainfall doesn’t increase with the elevation;
- extremes values (higher than 99,9 percentile) ap-
pear well correlated.
HYDROLOGICAL ANALYSES
The Campania area is characterised by a typical
Mediterranean climate, with dry season between June
and September and wet season from October to May.
The storms that occur at the beginning of the rainy sea-
son (Sept.-Oct.) can be characterized by a very high
peak of hourly intensity, possibly linked to heat and
local disturbance; the autumn-winter rainfalls appear
connected to storms characterized by either high hourly
intensity or long duration, and the spring season, on the
other hand, is characterized by more evenly distributed
rainfall (f
ioRillo
&w
ilson
, 2004).
Fig. 4 - Comparison between rainfall (2-days) recorded by rain gauges located at different elevation. a) Quindici and Piani
di Prato (3,1 km spaced), period 1998-2008; b) Lauro and Pozzo S.Romano (3,5 km spaced), period 1971-1977
background image
RAINFALL INITIATION OF DEBRIS FLOWS IN CAMPANIA (ITALY): A TWO-PHASE ANALYSIS
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
57
cover is reduced to its minimum level (θ
min
).
To increase the water volumetric content from θ
min
up to θ
max
, an amount of 240 mm of water is needed,
corresponding to soil storage capacity, m. Due to eva-
potranspiration processes, this amount of water corre-
sponds to an higher amount of rainfall, up to 400-500
mm, depending on the its temporal distribution.
In order to examine the antecedent soil moisture
conditions before several important storms, the path of
the soil moisture was computed from daily rainfall and
temperature, and a hydrological balance of the pyro-
clastic cover (f
ioRillo
& w
ilson
, 2004).
Table 3 shows the main storms (2 days cumu-
lative rainfall) extracted from a historical series of
the Lauro rain gauge (Quindici rain gauge after May
1998), for which antecedent volumetric water con-
tent, θ, was calculated.
The value
P is the minimum rainfall needed to
reaches
θ
max
, and it depends on the soil moisture con-
ditions and thickness of soil involved as below speci-
fied:
P=H (
θ
max
-
θ
)
The value
P makes up the minimum quantity
to be subtracted from Pstorm since runoff, especially
during high intensity rainfall, leads to an increase of
such an amount. Thus, the excess of rainfall, P
exc
use-
ful to induce positive pore pressure is given by:
P
exc
= P
storm
-
P
Table 3 shows that at the Lauro rain gauges the
highest Pexc values refer to the storms of May 1998
and Dec 2004. The other storms were less intense or
occurred in conditions of
θ
<
θ
max and show up lower
values for P
exc.
The storm of May 1998 is the powerful in term of
rainfall exceed, P
exc
, but other recent storms (Decem-
ber 2004 and March 2005) appear intense and will be
analyzed in the following section.
PHASE II: POSITIVE PORE PRESSURE AND
CONDITIONS LEADING TO DEBRIS FLOw INI-
TIATION
Detailed in situ measurements (P
ieRson
, 1980, s
i
-
dle
& s
wanston
, 1982; R
eid
et alii, 1988; J
oHnson
&
s
itaR
, 1990; w
ilson
& w
iezoReCk
, 1995; f
annin
&
J
aakkola
, 2000) demonstrated that positive pore pres-
sures develop during heavy storms.
w
ilson
& w
iezoReCk
(1995) noted that positive
pore pressures might be highly transient, decreasing
roots reach depth between 80 and 120 cm in the Sarno-
Quindici area. Plan roots are unable to cross the pum-
ice layers (Fig. 2); these levels act as a drainage sheet,
and break the root grooving deeply. As consequence,
the evapotranspiration processes occur prevalently in
the upper part of the pyroclastic mantle, deeply lim-
ited by the first pumice level. In situ suction measure-
ments (v
eRsaCe
et alii, 2005) highlight the increasing
of the suction toward summer, the decreasing towards
autumn; below pumice level the suction tends to be
constant during the year. For a thickness of 100 cm
of the zone subject to evapotranspiration processes,
and based on the technical characteristics of the soil
mantle, the following parameters have been estimated
(f
ioRillo
& w
ilson
, 2004):
- soil storage capacity, m, 240 mm;
- (volumetric) water content at field capacity, θ
max
, 51%
- minimum (volumetric) water content reached at the
end of dry season (summer), θ
min
, 27%;
- thickness of the soil, H, 1000 mm.
The soil generally reaches peak moisture (θ
max
) by
December and remains at or near that level until April.
During this period, any additional rainfall percolates
down, recharging the water table of the carbonate aq-
uifer, or runs off as surface stream flow. Between July
and September, the moisture content of the pyroclastic
Tab. 3 - Some hydrological parameters of the main storms
(Lauro/Quindici rain gauge) occurred in the area
analysed: P
storm
, 2-days cumulative rainfall of the
storm; θ, antecedent soil moisture;∆P, minimum
rainfall needed to reachθ,
max
; P
exc
, rainfall excess
and useful to induce positive pore pressure
background image
FRANCESCO FIORILLO & FRANCESCO M. GUADAGNO
58
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
quickly after the cessation of heavy rainfall. This could
make it difficult to measure or observe positive pore
pressures without automatic data recording of rapid-
response piezometers. Furthermore, in hillslopes with
stratified and/or lenticular soil layers with different
hydraulic conductivities, higher pore pressure will de-
velop in locations where more conductive zones pinch
out or are constricted (R
eid
et alii, 1988). J
oHnson
&
s
itaR
(1990) concluded that, because of the highly
variable nature of the pore pressure response, both in
space and time, and the close association with specific
characteristics of the rainfall record, a traditional hy-
drological model cannot completely explain the hy-
drologic response leading to debris-flow initiation on
the upper portion of the slope.
Although instrumental data about positive pore
pressures during debris flow initiation have not yet
been collected in the Campania pyroclastic mantle, it
is probable that positive pore pressures do form during
intense rain storms (f
ioRillo
& w
ilson
, 2004). Be-
cause of the high permeability of the Campania pyro-
clastic cover and the steep hill slopes, however, possi-
ble positive pore pressure would rapidly decrease after
each storm. Below, the role of positive pore pressure
has been evaluated by the "leaky barrel" model de-
scribed by w
ilson
& w
iezoReCk
(1995). This model
cannot explain the real hydrologic phenomena into the
soil mantle, but it is useful to evaluate the power of
each storm in function of its rainfall time-distribution.
The LB model describes the drainage character-
istics of a hydrological system, such as a slope, by
introducing a new parameter, the drainage coefficient,
k
d
, with dimensions of inverse time (hour
-1
). During a
storm, the amount of retained rainwater, Z, increases
or decreases, depending on the balance between rain-
fall intensity, I(t), and drainage rate, K
d.
Z. The net
rate of change in the water level retained in the system
may be expressed as a linear first-order differential
equation (w
ilson
& w
iezoReCk
, 1995):
dZ/dt = I(t) - k
d
Z
f
ioRillo
& w
ilson
(2004) compared the main
storms occurred in Campania using hourly rainfall
data, and found the minimum value of retained rain
water, Z
t
, needed to induce a failure. Here, the analy-
ses has been extended to further 10-years records of
rainfall to verify the model used. As shown in Tab.3,
during this recent decade, the storm of December
2004 and March 2005 are comparable to May 1998
storm. Figure 5 shows the values of the retained rain-
water, Z, induced by the storm of May 1998 and De-
cember 2004. The value of K
d
was deduced observing
the starting time of landslide initiation (gray bar of
Fig.5a). In this time, a higher value of Z must not have
been reached previously, nor exceed after all landslide
initiations. The minimum retained rainfall needed to
debris flows initiation, Z
t
, has been fixed at 75 mm for
Lauto rain gauge gauge and for K
d
=0,05 (Fiorillo &
Wilson, 2004). This value has not been reached during
the December 2004 and March 2005 storms (Fig.5b-
c), which failed to trigger debris flows in the Sarno
area, instead, the 4 March 2005 storm induced a debris
flows in Nocera, where it was more intense (Tab. 1)
DISCUSSION
Based on the historical rainfall data (daily and
hourly data) and the occurrence of the main debris
flow events since 1954, the hydrological conditions
leading to debris flows in Campania can be connected
Fig. 5 - Lauro rain gauge: hourly rainfall of main storms
of the 1998- 2008 period, and amount of rainfall
retained, Z, computed by Leaky Barrel Model, for
different kd values. The March 2005 and Decem-
ber 2004 storms didn’t reached the threshold Zt,
and failed to trigger debris flows in the Sarno
area. Gray bar indicates time of failure during the
May 1998 storm
background image
RAINFALL INITIATION OF DEBRIS FLOWS IN CAMPANIA (ITALY): A TWO-PHASE ANALYSIS
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
59
capacity (Fig. 6). Initial horizontal segment indicates
that antecedent rainfall and temperature conditions
are able to maintain field capacity also considering a
value of the soil moisture storage capacity, m, higher
than 240 mm. Whereas, decreasing trend indicates a
diminution of the soil moisture condition, which re-
duces the “power” of the storm, in term of P
exc.
The value of P
exc
of the May 1998 storm remains
constant up to a value of the soil moisture storage
capacity, m=480 mm. Also the May 1963, February
1984 and March 2005 storms present similar path, but
they were characterised by minor value of rainfall ex-
cess, Pexc. The March 2005 storm was more intense
at Nocera rain gauge (167 mm, Tab.1) and locally in-
duced some debris flows. The rainfall excess of other
storms drop rapidly for values of the soil moisture
storage capacity, m, higher than 240 mm, indicating
that storm rainwater was partially retained by the py-
roclastic mantle as retention water.
ConCludIng remarKs
This study extends the hydrological analyses on
debris flows in the Campania area carried out by f
io
-
Rillo
& w
ilson
(2004), integrating data with further
10-years of records.
Two main aspects have been discussed:
- the possible rainfall amount in the high elevation
zones during the May 1998 storm;
- the role of the antecedent rainfall controlling the
amount of the rainfall excess during intense storm.
The first point has been analysed by high eleva-
tion data available after the May 1998 storm, and by
another rain gauge during the seventieths. The analy-
sis shows that rainfall doesn’t increase with the eleva-
tion, and the singularity of the May 1998 storm has to
found in the characteristics of the antecedent rainfall
to two main principal factors:
-
a long antecedent period of rainfall, which
recharge the pyroclastic mantle after the water deficit
accumulated during the dry and summer;
-
a storm characterised by high intensity in
term of retained rainwater.
Conditions of soil moisture close to field capac-
ity are reached during the winter season, and can be
maintained also during the spring season. Minor is the
soil moisture conditions of the pyroclastic mantle pre-
vious the storm, and higher has to be the intensity of
the storm to induce debris flows.
The storm of May 1998 and October 1954 reflect
the two extreme records which lead to debris flow ini-
tiation.
The storm of October 1954 occurred after a long
dry season (∆P, the minimum rainfall needed to reach
θ
max
was at the maximum value, 240 mm) and was in-
tense in the Salerno Area (fig.3) and induced rainfall
excess, P
exc
.
Viceversa, the storm of May 1998 wasn’t particu-
larly intense. It singular characteristics appear con-
nected to the previous hydrological conditions. To
stress this point some considerations will be given on
the capacity of the soil to retain water. We have fixed
a value of soil moisture storage capacity of 240 mm,
which refers to 1 m thick evapotranspiration zone.
However, soil moisture storage capacity cannot be as-
sumed as a constant, depending mainly on the thick-
ness of soil, and in particular on the depth of the first
pumice level.
Higher is the soil moisture storage capacity, high-
er will be the amount of rainfall needed to reach the
field capacity. Besides, the effects of the storm tends
to increase as the antecedent soil moisture increase.
As consequences, any specific storm, character-
ised by a definite hourly rainfall distribution, induce
different pore pressure into the soil, in function of the
soil moisture storage capacity of the soil.
In situ thickness measurements of the pyroclastic
mantle, carried out in the scar zones, are somewhere
higher than 1 m, and can reaches a value up to 2 m. If
the thickness of the pyroclastic mantle subject to eva-
potranspiration is considered higher than 1 m, a higher
values (>240 mm) of the soil moisture storage capac-
ity, m, has to be assumed. Based on this hypothesis,
the value of the rainfall excess, P
exc
, has been re-com-
puted for different values of the soil moisture storage
Fig. 6 - Rainfall excess, Pexc, as function of soil moisture
storage capacity, m, for the main storms occurred
at Lauro rain gauge extracted from table 3
background image
FRANCESCO FIORILLO & FRANCESCO M. GUADAGNO
60
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
The spatial distribution of different values of the
soil moisture storage capacity can have a strong control
on the location of the landslide initiation. This suggests
that landslide initiation in Campania is also controlled
by the characteristics of the storm and by antecedent
hydrological conditions, as well as geomorphological
and stratigraphical features of the slopes.
ACKNOWLEDGEMENTS
The authors are grateful to “Centro funzionale per
la previsione meteorologica ed il monitoraggio meteo-
idro-pluviometrico” of Campania, Italy, for the hydro-
logical data of the 1998-2008 period.
and soil moisture storage capacity.
The second point has been discussed by the Fig.6:
zones characterised by high values of the soil mois-
ture storage capacity are sensible to powerful storm
only after abundant antecedent rainfall recharge.
These characteristics help to understand the cata-
strophic effects induced on the slopes by the May
1998 storm, although it had unexceptional character-
istics in term a total amount of rainfall. The singular
characteristics of the hydrological condition before
the May 1998 storm lies in the particular distribu-
tion of the antecedent rainfall, in a way that soil could
maintain field capacity condition up to a high value of
the soil moisture storage capacity.
REFERENCES
b
RaCa
G. & o
noRati
G. (1998) - Relazione sull'evento idrogeologico del 4, 5 e 6 maggio 1998 in Campania - Dipartimento
per i Servizi Tecnici Nazionali, Servizio Idrografico e Mareografico Nazionale; Rome, 30 September 1998. Unpublished
technical report.
b
RanCaCCio
l., C
inQue
a., R
usso
f. & s
Gambati
d. (2000) – Le frane del 5-6 maggio 1998 sul gruppo montuoso Pizzo d’Alvano
(Campania): osservazioni geomorfologiche sulla loro distribuzione e sulla dinamica delle connesse colate. Quaderni di
Geologia Applicata, Pitagora ed., Bologna, 7(1): 5-36.
b
RanCaCCio
l., C
inQue
a., & s
GRosso
i. (1978) – L’analisi morfologica dei versanti come strumento per la ricostruzione degli
eventi neotettonici - Memorie Soc. Geol. It., 19: 621-626.
C
alCateRRa
d., P
aRise
m., P
alma
b., P
elella
l. (1999) - The May 1998 landslides event in Campania (Southern Italy): inven-
tory of slope movements in the Quindici area - Proc. Symp. Slope Stability Engineering; Y
aGi
y. & J
ianG
eds
., Balkema,
Rotterdam, 1361-1366.
C
alaCateRRa
d. & s
anto
a. (2004) - The January 10, 1997 Pozzano landslide, Sorrento Peninsula, Italy. Eng. Geol., 75(2):
181-200.
C
ivita
m.,
de
R
iso
R., l
uCini
P. & d’e
loGio
n
ota
(1975) - Studio delle condizioni di stabilita` dei terreni della Penisola Sor-
rentina (Campania). Geol. Appl. Idrogeol., 10: 129-188.
C
Rosta
G .b.& d
al
n
eGRo
P. (2003) - Observations and modelling of soil slip-debris flow initiation processes in pyroclastic
deposits: the Sarno 1998 event. Nat. Hazards Earth Syst. Sci., 3: 53-69.
C
Ruden
d.m. & v
aRnes
d.J. (1996) - Landslide Types and Processes - In: Landslides: Investigation and Mitigation. e
d
. t
uRneR
a.R. & s
CHusteR
R.l., Sp. Rep. 247, Transportation Research Board, National Research Council, National Academy Press,
Washington D.C., 36-72.
d
el
P
Rete
m. & d
el
P
Rete
R. (2002) - Activation Mechanisms of Campanian debris flows of May 1998, in Southern ltaly - 9th
Int. Conf. IAEG ,16-20
th
September 2002, Durban (South Africa), 1889-1898.
d
el
P
Rete
m., G
uadaGno
f.m. & H
awkins
a.b. (1998) - Preliminary report on the landslide of 5 May 1998, Campania, southern
Italy. Bull. Eng. Geol. Env., 57: 1998, 113-129.
d
e
R
iso
R., b
udetta
P., C
alCateRRa
d. & s
anto
a. (1999) - Le colate rapide in terreni piroclastici del territorio campano
Conf. Previsione e prevenzione di movimenti franosi rapidi, 17-19 June 1999, Trento; Associazione Georirorse e Ambiente,
Trento; 133-150.
e
sPosito
l. & G
uadaGno
f.m. (1998) - Some special geotechnical properties of pumice deposits. Bull. Eng. Geol. Env., 57:41-50.
f
annin
R.i. & J
aakkola
l
. (2000) - Piezometric observation: Implication of debris flow initiation on forested hillslopes - Proc.
of the 8
th
Int. Symp. on Landslides held in Cardiff on 26-30 June 2000- Landslides in research, theory and practice, Thomas
Thelford ed., 2: 537-542.
f
ioRillo
f., G
uadaGno
f.m., a
Quino
s. & d
e
b
lasio
a. (2001) - The December 1999 Cervinara landslides: further debris flow
background image
RAINFALL INITIATION OF DEBRIS FLOWS IN CAMPANIA (ITALY): A TWO-PHASE ANALYSIS
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
61
in the pyroclastic deposits of Campania (Southern ltaly). Bull. Eng. Geol. Env., 60: 171-184.
f
ioRillo
f. & w
ilson
R.C. (2004) - Rainfall induced debris flows in pyroclastic deposits, Campania (southern Italy). Engineer-
ing Geology, 75(3-4): 263-289
G
uadaGno
f.m., f
oRte
R., R
evellino
P., f
ioRillo
f. & f
oCaReta
m. (2005) - Some aspects of the initiation of debris avalanches
in the Campania Region: the role of morphological slope discontinuities and the development of failure - Geomorphology,
66: 237-254.
G
uadaGno
f.m & P
eRRiello
z
amPelli
s. (2000) - Triggering mechanisms of the landslides that invested Sarno, Quindici, Siano
and Bracigliano (Southern ltaly) on May 5-6, 1998. Proc. of the 8
th
Int. Symp. on Landslides held in Cardiff on 26-30 June
2000- Landslides in research, theory and practice, T. Thelford ed., 2: 671-676.
G
uadaGno
f.m., s
CaRasCia
m
uGnozza
G.& m
aRtino
s. (2003) - lnfluence of man-made cuts on the stability of pyroclastic cov-
ers (Campania - Southern ltaly): a numerical modelling approach. Env. Geol., 43: 371-384.
H
unGR
o., e
vans
s., b
ovis
m.i. & H
utCHinson
i.n. (2001) - A review of classification of landslides of the flow type - Env. &
Eng. Geoscience, 7: 221-238.
J
oHnson
k.a. & s
itaR
n. (1990) - Hydrologic conditions leading to debris-flow initiation. Can. Geotech. J., 27: 789-801.
m
azzaRella
a., m
aRtone
m. & t
RanfaGlia
G. (2000) - Il recente evento alluvionale del 4-5 maggio 1998 nel Sarnese e il deficit
risolutivo della rete pluviografica. Quaderni di Geologia Applicata, 7 (2): 203-211.
m
ele
R. & d
el
P
Rete
s. (1999) - Lo studio della franosità storica come utile strumento per la valutazione della pericolosità da
frane. Un esempio nell'area di Gragnano (Campania)- Boll. Soc. Geol. It., 118,91-111.
o
livaRes
l., a
ndReozzi
e., d
amiano
b., a
voRio
a
& P
iCaRelli
l. (2003) – Hydrologic response of a steep slope in unsatured
pyroclastic soils. Int. Conf. “Fast Slope Movements Prediction and Prevention for Risk Mitigation” Naples, May 11-13,
2003, l. P
iCaRelli
ed
., 1: 394-397.
P
aResCHi
m.t., f
avalli
m., G
iannini
f., s
ulPizio
R., z
anCHetta
G. & s
antaCRoCe
R. (2000)- May 5, 1998, debris flows in circum-
Vesuvian area (southern Italy): insights for hazard assessmen. Geology, 28(7): 639-642.
P
enta
f., l
uPino
R., C
aPozza
f. & e
su
f. (1954) - Effetti dell'alluvione del 26 ottobre 1954 nel Salernitano. Rivista Italiana di
Geotecnica, 6: 245-258.
P
ieRson
(1980) - Piezometric response to rainstorms in forested drainage depression. Journal of Hydrology, 19: 1-10.
R
eid
m.e., n
ielsen
H.P., d
Reiss
s.J. (1988) - Hydrologic factors triggering a shallow hillslope failure. Bull. of the Assoc. of
Eng. Geologists, XXV(3): 349-361.
R
evellino
P., H
unGR
o., G
uadaGno
f.m., e
vans
s.G. (2004) – Velocity and runout simulation of destructive debris flows and
debris avalanches in pyroclastic deposits, Campania region, Italy. Env. Geology, 45: 295-311.
R
osi
m. & s
bana
a. (Eds.) (1987) - Phlegrean Fields. Quaderni della Ricerca Scientifica, 9, C.N.R., Rome, 168 pp.
S.I.M.N. (1954) - L'alluvione del Salernitano (25-26 ottobre 1954). Annali idrologici, Servizio Idrografico e Mareografico di
Napoli, parte Il.
s
antaCRoCe
R. (Ed.) (1987) - Somma-Vesuvius. Quaderni della Ricerca Scientifica, 114, C.N.R., Rome, 243 pp.
s
idle
R.C. & s
wanston
d.n.(1982) - Analysis of a small debris slide in coastal Alaska. Can. Geotech. J., 19: 167-174.
t
eRRibile
f., b
asile
a., d
e
m
asCellis
R.,
di
G
ennaRo
a., m
ele
G. v
inGiani
s. (2000) - I suoli delle aree di crisi di Quindici e
Sarno: proprietà e comportamenti in relazione ai fenomeni franosi del 1998 - Quaderni di Geologia Applicata, 7 (1): 59-95.
v
eRsaCe
P., C
aRuso
P., C
assetti
m. & C
aPPaRelli
G. Il presidio territoriale e la gestione dell'emergenza a Sarno. Workshop
“La mitigazione del rischio da colate di fango a Sarno e negli altri comuni colpiti dagli eventi del maggio 1998”, May 2005,
Naples: 55-78.
w
ilson
R.C. & w
iezoReCk
G. f. (1995) - Rainfall thresholds for the initiation of debris flows at La Honda, California. Env. &
Eng. Geoscience, 1(1): 11-27.
Statistics