modelling the timing and location of shallow landslides ... · rainfall is an important factor for...
TRANSCRIPT
1.Synthetic rainfall events were generated corresponding to selected observed extreme
rainfall totals of given return period (RP), for a range of durations (1-day to 5-day)
• A Generalised Extreme Value distribution was fitted to observed annual maxima
data using L-moments. 207
207
196
147
109
218
210
204
159
233
237
231
224
208
167
237
237
225210
169
237237226214175
0
1
2
3
4
5
6
7
0 1 2 3 4 5 6
Caine 1980 - Worldwide
Innes 1983 - Worldwide
Crosta & Frattini 2001 - Worldwide
Crosta & Frattini 2001 - Southern Alps
Literature
values of
landslide
thresholds
Number of simulated landslides
Estimated curves for constant
number of simulated landslides
{
Results
Acknowledgements: This work forms part of the LESSLOSS FP6 project (www.lessloss.org) (GOCE-CT-2003-505448) and builds upon the DAMOCLES
FP5 project (EVG1-CT-1999-00007).
References: � Caine N. 1980, The rainfall intensity-duration control of shallow landslides and debris flows. Geografiska Annaler, 62 A, 1-2,23-27 � Crosta
G. B., Frattini P., 2001. Rainfall thresholds for triggering soil slips and debris flow. In Mugnai A., Guzzetti F., Roth G (eds) Mediterranean Storms, 2nd
Plinius Conference, pp 463-487 � Innes J. L., 1983. Debris Flows. Progress in Physical Geography, 7, 469-501
Focus Area & Input data
The Pioverna (Valsassina) catchment (160 km2) & Esino catchment (20 km2) are
located in the pre-Alps of the Lecco province in Northern Italy. The rivers discharge into
Lake Como (Lario). The area receives a mean annual rainfall of 1542 mm.
C. Isabella Bovolo and James C. Bathurst
School of Civil Engineering and Geosciences, University of Newcastle upon Tyne, [email protected], [email protected]
0
5
10
15
20
0 2 4 6 8 10
12
14
16
18
20
22
24
"Decreasing"
Rainfall
"Increasing"
Rainfall
0
10
20
30
402-year RP
5-year RP
10-year RP
25-year RP
50-year RP
"Peak" Rainfall
"Constant"
Rainfall
0
20000
40000
60000
80000
100000
120000
140000
0 20
40
60
80
100
120
140
160
Modelling the Timing and Location of Shallow Landslides and
Debris Flows using the SHETRAN Model
Introduction
Rainfall is an important factor for triggering landslides but its effect is poorly understood.
The aim of this work is to investigate the occurrence, and the associated sediment
release, of shallow landslides and debris flows at the river basin scale as a function of
rainfall return period. The timings and locations of landslides were simulated in
response to rainfall patterns of different combinations of intensity and duration. The
simulations were carried out using the SHETRAN landslide model.
Method
Soilcl20123No Data
Forests
Pastures and
meadows
Rocky outcrops
Conclusions
• The effects of different rainfall properties on simulated shallow landslide incidence and
sediment yield were explored.
• The relationship between the simulated number of landslides and rainfall event intensity
and duration was found to be similar to that in the literature. The upper bound
simulations were found to be too high but the lower bound simulations were found to be
consistent with Innes (1983) and Crosta & Frattini (2001).
• The timing and occurrence of landslides was shown to be influenced by rainfall
distribution pattern as well as magnitude. The effects of antecedent conditions should
now be explored.
• These results should help to develop forecasting, mitigation and hazard zonation
strategies.
Combined Sediment Yield for
Pioverna & Esino River outlets
Tota
l Sedim
ent Yie
ld (to
nnes)
Rainfall event intensity (mm/day)
Simulated number of landslides compared to published
zero-landslide thresholds for lower bound simulations
Avera
ge rain
fall
event in
tensity (m
m/h
r)
Rainfall duration (days)
Number of landslides associated with different return periods
for the upper bound simulations
Precipitation, river discharge and number of landslides
for a 1-day rainfall event with a 2 year return period
Num
ber of la
ndslid
es
Pre
cip
itation (m
m)
Hours
Location of landslides for
rainfall events of 1 day
duration & 2 year return
periodLandslide location (lower bound)
Landslide location (upper bound)
River
Evapotranspiration
Canopy interception
Root zone
Landslides,
erosion & sediment
transport
3D
Variably
saturated
subsurface
flow model
Snowmelt
Overland
& channel
flow
SHETRAN V4
The SHETRAN landslide model
• Landslide occurrence is determined as a function of the time- and space-varying soil
saturation conditions simulated by SHETRAN using infinite-slope, factor-of-safety
analysis.
• The SHETRAN landslide model is relevant to catchment scales of up to 500 km2.
• SHETRAN can predictively examine the impacts of possible future changes in climate
(including rainfall characteristics) and can therefore be used in hazard assessment and
as a tool for developing mitigation strategies.
The landslide component
uses the topographic index
to link the SHETRAN grid
resolution (typically 100m –
2km), at which the basin
hydrology and sediment
yield are modelled, to the
landslide component’s sub-
grid resolution (typically 10 –
20m), at which landslide
occurrence and erosion are
modelled.
• SHETRAN is a physically based, spatially distributed, integrated surface / subsurface
modelling system for water flow and sediment transport in river catchments.
• The landslide component models shallow landslide occurrence and the resulting
sediment yield at the catchment scale.
Elevation (m)
Esino river
Pioverna river
• 1- 5 day rainfall events, corresponding
to selected return periods, were
disaggregated to hourly resolution.
• Four rainfall disaggregation patterns
were evaluated with equal total
precipitation but variable hourly
intensity, called: Peak, Constant,
Increasing and Decreasing.
2.Uncertainty was allowed for by setting upper
and lower bounds on key model parameters.
3.The rainfall events were applied to the
SHETRAN model which simulated landslides
and the resulting sediment yield.
4.The results were analysed to investigate
relationships between rainfall properties and
the characteristics of the modelled landslides
and their sediment yield.
Rainfall disaggregation pattern.
1-day example shown below.
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
0 5 10 15 20 25 30 35 40 45 50
1 day events
2 day events
3 day events
4 day events
5 day events
Rainfall Return Period
Simulations without use of the landslide component
are the same as for the Lower Bound simulations
Upper Bound
Lower Bound
1 day events
2 day events
3 day events
4 day events
25 year RP
50 year RP
5 day events
2 year RP
5 year RP
10 year RP
Dis
charg
e (m
m/s
)
Num
ber of la
ndslid
es (/1
0)
Time (hours)
100 200150
SHETRAN grid resolution = 500 m.
Landslide grid resolution = 20 m.
Soil class and
sand content (%)
(32 – 44)
(48 – 54)
(60 – 74)
Stream – aquifer
interactions
!!!
!!! !!!!!!!!
!!!! ! !!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!! !!! !!!!! !!!!!!!!!!!! ! !! !!!! !!! !!!!!! !!!!! !!!! !!!!! !!!!! !!!!! ! !!!!! !!!!! !!! !!!!!! !!!!!!! ! !!! !!! !!! !! !!!! !!!! !!!!!!! !!!!!!!!!!!!!!! ! !!!!!!!!!!!! ! !! !! !!!!!!!! !!!! !!!!!! !!!!!! !!! !!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!! !!! !!!!!!!!!!!! !!! ! !!!!!!! !!!!!!! !!!! !! !!!!!!!!!!!! ! !! !!!!!! !!!!!!! !! !!!!!! !!! !!! !!! !!!!!! ! !!!!!! !! !! !! !!!!!! !!!! !!! ! !!!!! ! !!! !!! ! !! !! !! ! !!!! ! !! !!!!!!! ! !! !! !!!! !!! !! !!!!!!! !!!! !! !!!!!! !! !! !! !! !!!!! !! !!!!! !! !!!!!! ! !! !!!!!!!!! ! !!!!!! !! ! !!!!!!!! ! !!!!!! !! !!!!!!! ! !!! !!!!!!!! ! ! !!!!! !!!!! !!!!!!! !!! !!!! !!!!!!!!!!!!!!!! !!!!!!!! !!!! !!! !!!!!!!! !!!! !!!!! !!!! !!!!! ! !! ! !!! ! ! !!!!!!! !!!!!!!! !! !!! ! !!! ! !! !! !!! !! !!!! !!!!! ! !!!!! ! !!!!!! !!!
!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!! ! !!!!
!!!!
! !!!!!!!
!! !! !!!!!!!! !! !!!! !!!! !!! !!! !! ! !!! ! !!! !! !!!! !!!!!!!! !! ! !!!!!! !! !!! !!!!! !!!! !!!!!! !!!!!!!!!!!!!!!!!!!!!!!!! !!!!!! !!!! !!!!! !!!! !!!!!!! !! !!! ! !!!!!! ! !!!!!!!!
!
!!!!!!!!! !!!!!!!! !! !!!!!!! !! !!!!!!!! ! ! !!!! !!!! ! !!!!!!! !!!!!! !!! !!!!!!! !!!!!!!!!!!! ! !! ! !!!! ! !!!!! !!!!! !! !!!!! !! !! !!!!! !! !!!!! !! !!! !!! ! !!! !!!!! ! !!!! ! !!!! !! ! !!! !! !!!! ! !!! !!!!! ! !!!!!!! !!!!! !!!!!!!!! ! !! ! !!!
!! !!! ! ! !! !!!! ! !!!! ! ! !!!! ! !! !!!!!!! ! ! !!! !! ! ! !!!! !! !! !! ! !!! !! ! !! !!! !! !!!! !!!!!!!!!! !!!!!!!!!! !!!!!!!!!!!
!! !! !!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !! !!!!!!!!!!!!!! !!!!!!! !!!! !! !!!! !! !!!!!! ! !!!! !!! ! !!! !!!!!! !!! ! !!!! !! !!!!! !! !! !!!!!!!!!!! !!! !!!!!!! !!!!!!! !!! !!!!!!! !!!!!!! ! !!!!!!!! !!!! !!!! ! !!!!! !!! ! !!! !!!!! !!!! !!!!! !!!!! !! !!!! !! ! !! !! !! ! ! !!!!!!!!!!!!!!!!! !!!!!!!!!! !!!! !!!!! !! ! ! !! ! !! !! !!!! !!!! !!!!!! ! !!!! ! ! ! !!!!!!! ! !! !!!!!! ! ! ! !! !!!!!!!! !! !!!!!!!!!!!! !!!!!!!! !!!!!!!!!!! ! !!!!!!! !!!! !!!!!! !!!!!! ! !! ! !!! !!! !!!! !! ! !!! !!!! !! ! !!!!!!!! !!!!!!!!! ! !!!!!!! !!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!! !!!!!!!!!!!
! !!! !! !!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ! !!!!!!!!! !!!! !!!! !!!! !!!!! !!!!! !!!! !!! !! !!! !!!! !!!!! !!!!!! !!!!! !!!! !!!!!!!!!! !!!!!!!! !!! !!!! !!!! !!! !! ! !!! !! !!!!! !!!!!!!!!!!! !!!!!!!! ! !!!! !!! !!! !! !! ! !!!! ! ! !! ! !!!!!!! !! !!!! !!! ! !!!!! ! !!!!!!! !! !!!!! !! !!!! !!! !!!!!! !!!! !!
! ! !!! !!!!! !!! !!!!!!! !!!!!!!!!!!!! !!!!!!!! !!!! !!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!! !!!!!!!!!! !!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!! !!! !! !!!!!!!!!!
!!!!
!!!!!!!!!!!!!!!!! !!! ! !! !
!!!! !!!!!! !!! !!!!! !!
!
!!!! !!! !!! !! !!!!!!!!!! !!!!!!!!
!!!!
!!
!!!!! !!! !!!!!!!!!!!!
!!! !
!!
!!