See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268814755

The Application of the Erosion Potential Method to Alpine Areas:

Methodological Improvements and Test Case

Conference Paper · January 2014

DOI: 10.1007/978-3-319-09054-2_73

CITATION READS

1 669

3 authors, including:

Luca Milanesi Marco Pilotti

Università degli Studi di Brescia Università degli Studi di Brescia
22 PUBLICATIONS   217 CITATIONS    84 PUBLICATIONS   1,000 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Flood hazard mapping View project

Lake Garda limnological monitoring View project

All content following this page was uploaded by Luca Milanesi on 03 May 2015.

The user has requested enhancement of the downloaded file.

 The Application of the Erosion Potential
Method to Alpine Areas: Methodological 73
Improvements and Test Case

Luca Milanesi, Marco Pilotti, and Alberto Clerici

Abstract
The Erosion Potential Method is a valuable instrument for the estimate of the mean annual
sediment yield at the basin scale. The goal of this study was to demonstrate its applicability in
Alpine basins with relevant presence of areas periodically covered by snow and ice. To this
purpose, in order to account for the presence of frozen soil, the application of the model to
the thawing periods only may be advisable. A distributed approach for the application of the
method was suggested in order to reproduce the features of the original calibration,
characterized by field data and experiments at the parcel scale. Finally, the model was tested
on two Italian Alpine basins characterized by the presence of glaciers and perennial snow and
that are drained by a barrage for hydropower purposes. The results of the model were
compared to data of suspended sediment measured at the barrage by the plants manager. The
good fit between the measured data and the model outputs shows the reliability of the modified
model in Alpine areas.

  
Keywords
Erosion potential method Reservoir sedimentation Sediment yield Soil erosion

the sediment budget. The hydraulic approach deals with the

73.1 Introduction
The assessment of soil erosion intensity is of primary
importance for land use planning, soil conservation, pollu-
tion control and water resources management (Ranzi et al.
2012). For instance, 35 % of the original reservoir storage
capacity at the World level is now occupied by sediments
(Basson 2010). For these reasons many researches were
prompted in order to develop instruments for the estimate of
L. Milanesi (&)  M. Pilotti  A. Clerici
Department of Civil Engineering, Architecture, Land,
Environment and of Mathematics – DICATAM,
Università degli Studi di Brescia, Brescia, Italy
e-mail: luca.milanesi@unibs.it
problem on the basis of the transport capacity of the river in
each cross section, disregarding the availability of sediments
at the basin scale. On the contrary, conceptualized models
evaluate the intensity of erosion processes at the basin or the
parcel scale with different time horizons. Among this class of
models, on the basis of their approach, the most common
methods can be classified as empirical, conceptual and
physically based.
The Erosion Potential Method (Gavrilovic 1988), in the
following also indicated as “EPM”, is an empirical semi
distributed model to estimate the mean annual volumes of
soil erosion and sediment yield at the basin scale. It com-
bines in a simple structure all the most statistically signifi-
cant parameters controlling soil particles detachment and
transport. Although calibrated for the Dinaric Alps, several
applications to the Italian and Swiss Alps can be find in the
literature. Moreover, this model is widely used for soil
erosion mapping in semi arid regions.

G. Lollino et al. (eds.), Engineering Geology for Society and Territory – Volume 3, 347
DOI: 10.1007/978-3-319-09054-2_73, © Springer International Publishing Switzerland 2015
348
73.2 Model Description
The specific mean annual soil erosion Wsp [m3/km2/y]
depends on the mean cumulative annual rainfall H [mm], on
the erosion coefficient Z [-] and on the parameter T [-], that is
a function of the mean annual temperature t [°C]:
pffiffiffiffiffi
Wsp ¼ T  H  p  Z3 ð73:1Þ
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
t Cesare et al. 1998). In particular, from November to April
T¼ þ 0:1 ð73:2Þ
10
 pffi 
Z ¼ Xa  Y  / þ i ð73:3Þ
The erosion coefficient is a combination of the soil pro-
tection coefficient Xa [-], of the parameter Y [-] describing
erodibility, of the active erosion processes coefficient φ [-]
and of the mean slope of the basin i [m/m]. The parameters
in Eq. (73.3) are computed by means of tables proposed by
Gavrilovic (1988) and reviewed by several authors (e.g.,
Zemljic 1971; Globevnik et al. 2003; Fanetti and Vezzoli
2007). The sediment yield volume G [m3/y] depends on the
drained area of the basin F [km2] and on the retention
coefficient R [-], which represents the percent of eroded
sediments that reaches the outlet of the basin:
G ¼ Wsp  F  R ð73:4Þ
According to the original formulation, R depends on the
length L [km] of the main channel and on the mean relief of
the basin. Literature studies (e.g., Zemljic 1971) and tests
performed by the Authors showed that the original formula
of R by Gavrilovic (1988) may return values greater than 1,
that would imply a sediment yield greater than the erosion
volume. Hence, in this paper the alternative equation sug-
gested by Zemljic (1971) was implemented:
qffiffiffiffiffiffiffiffiffiffiffi P qffiffiffiffiffiffiffiffiffiffiffi
^ Li ^
OH i OH
R¼ ¼ D ð73:5Þ
ðL þ 10Þ F ðL þ 10Þ
where Ĥ [km a.s.l.] is the mean altitude of the basin, and Li
[km] is the length of the reaches of ith order.
73.3 Methodological Improvements to EPM
EPM was originally derived for the Dinaric Alps through
experimental analyses and field data at the parcel scale. The
modifications suggested in the following paragraphs wish to
improve the applicability of EPM in areas with different
climatic features from the Dinaric Alps. Moreover a novel
calculation approach is suggested in order to couple the
L. Milanesi et al.
application at the basin scale, while keeping the local details
of the processes.
Since Alpine catchments are often characterized by mean
annual temperature below 0 °C, the definition of the coef-
ficient T reveals some troubles due the negative radical in
Eq. (73.2). Moving from this mathematical constraint and
observing the physics of erosional processes, it may be
advisable to apply the EPM with reference to the period
when erosion rate and sediment yield are relevant (De
most of the areas experience completely frozen soil with low
discharge in the streams so that soil detachment and trans-
port is negligible. For this reason, in this paper EPM was
applied calculating the mean temperature and the cumulative
rainfall from May to October.
Although EPM was derived by experimental analysis at
the parcel scale, both lumped (e.g., Tazioli 2009) and dis-
tributed (e.g., Bemporad et al. 1997) approaches can be
found in literature. In this study EPM was applied in a dis-
tributed form through a properly developed algorithm in GIS
environment. This code was primarily used for the extraction
of topologic and geomorphological information from a
Digital Terrain Model (Pilotti et al. 1996) and then for the
local application of the model equations (Pilotti and Bacchi
1997). The algorithm moves from link to link within the
space filling drainage network and follows the same direc-
tion of the runoff on the surface. This approach was used to
assess the local production of sediment using a DTM with a
20 m cell resolution. Equations from (73.1) to (73.3) were
applied in a distributed way, estimating the cumulative
erosion along the drainage network and keeping into account
the spatial variability of the input parameters. To this pur-
pose, the input data were recorded on raster format and the
parameters were expressed linking the original tables of the
method through up-to-date classification criteria (Milanesi
et al. 2013).
73.4 Application to the Cedec and Frodolfo
Catchments
Two small catchments in Alta Valtellina (Northern Italy)
were studied in order to assess the performances of EPM in
Alpine areas. Most of the water of these basins, drained
respectively by the Cedec (Area 17.8 km2, mean slope
0.48 m/m, mean altitude 2,909 m a.s.l.) and Frodolfo (Area
29.1 km2, mean slope 0.49 m/m, mean altitude 3,014 m a.s.
l.) torrents, are stored in a small barrage at the confluence
(2,167 m a.s.l.) and then diverted for hydropower purposes.
These basins, whose maximum altitude is about 3,800 m a.s.
l., are characterized by the presence of glaciers and perennial
snow in the upper part. While the coarser sediment fraction
is mostly stopped at the barrage by trap basins, the fine
73 The Application of the Erosion Potential Method to Alpine Areas 349

Table 73.1 Main parameters involved in the EPM formulas for the studied basins
Parameter (%) Frodolfo basin Cedec basin
Corine land cover Bare rocks (3.3.2) 33.2 46.3
Glaciers and perpetual snow (3.3.5) 57.5 28.7
Natural grasslands (3.2.1) 0.1 1.3
Sparsely vegetated areas (3.3.3) 9.2 23.4
Geology Hard rock masses 4.5 4.3
Weak rock masses 27.7 22.0
Coarse soils 0.63 0.8
Poorly sorted soils 2.3 5.8
Fine and medium soils 19.7 50.6
Glaciers 44.8 16.2
Active erosion Rockfalls 1.9 4.5
Avalanches 3.7 4.4
Rill erosion 2.5 15.5

fraction flows towards the diversion structures. Both these
catchments show a very low vegetal coverage and a signif-
icant contribution in silt sediment production from the gla-
ciers. The most relevant active erosive processes in these
areas are rockfalls, avalanches and, on the case of the Cedec
basin, rill erosion. A mean annual temperature of 1.7 °C and
a mean annual precipitation of 944 mm were registered at the
barrage. In the period from May to October, these data
modify to 7.5 °C and 656 mm. Finally, the assessment of the
local temperature in each cell was based on a −0.004 °C/m
gradient, while the precipitation value was assumed constant
on the basin. Table 73.1 summarizes the data needed for the
application of EPM in the study area.
73.5 Results
The application of EPM led to a mean volume of eroded
sediments of 26,030 m3/y for the Frodolfo catchment and of
23,600 m3/y for the Cedec. Since these basins have an
estimated retention coefficient of 0.786 and 0.607 respec-
tively, their mean sediment yield, considering the full par-
ticle size spectrum, is about 20,500 and 14,300 m3/y.
Although for these basins the relative difference between the
results provided by a distributed algorithm and the ones of a
lumped approach is limited to 2.5 %, in several cases it may
be greater than 10 %.
The estimates obtained by the application of EPM were
compared to the measured turbidimetric data registered since
2009 by the plants manager at the barrage during spring and
summer periods. Only the instantaneous value of suspended
sediments is registered, while no information on the coarse
fraction are available. A mean cumulative annual volume of
suspended load of 18,000 m3/y was measured.
In order to make the sediment yield estimates comparable
with the measured data, 8 soil samples were collected in the
study area and their granulometric curves were measured.
This procedure allowed to identify an average content of fine
material (with mean diameter smaller than 0.15 mm) of
45 %. Then, the volume of fine sediment yield was estimated
by multiplying this percent by the whole sediment yield
volume. A final result of 9,200 m3/y for the Frodolfo basin
and of 6,400 m3/y for the Cedec catchment was obtained, for
a global estimate of 15,400 m3/y.
Despite its simplicity, EPM provides reliable sediment
erosion estimates and, on the basis of the suggested modifi-
cations, it may prove to be an useful instrument for the
assessment of the intensity of this phenomenon also in Alpine
areas slightly different from the original calibration environ-
ment. The modifications suggested for the application stage
reflect more accurately the calibration approach followed by
Gavrilovic (1988) in the definition of EPM parameters.
Acknowledgments We wish to thank A2A staff, especially Dr. Michele
De Censi, Eng. Ferdinando Bondiolotti and Eng. Sergio De Campo, for
their valuable suggestions and disposability in data collection activities.
This work has been partially developed within the EU Project
KULTURisk FP7, Grant Agreement no. 265280.
References
Basson G (2010) Tackling a growing problem. Addressing the issue of
sedimentation. Int Water Power Dam Constr 62(7):38–40
Bemporad GA, Alterach J, Amighetti FF, Peviani M, Saccardo I (1997)
A distributed approach for sediment yield evaluation in Alpine
regions. J Hydrol 197:370–392
De Cesare G, Beyer Portner NA, Boillat J-L, Schleiss A (1998)
Modelling of erosion and sedimentation based on field investiga-
tions in Alpine reservoirs of hydropower schemes. In: Sediment
transport in reservoirs
350
Fanetti D, Vezzoli L (2007) Sediment input evolution of lacustrine
deltas: the Breggia and Greggio rivers case study (Lake Como,
Italy). Quatern Int 173–174:113–124
Gavrilovic Z (1988) The use of an empirical method (Erosion Potential
Method) for calculating sediment production and transportation in
unstudied or torrential streams. In: International conference on River
Regime, Wallingford, England, 18–20 May 1988, pp 411–422
Globevnik L, Holjevic D, Petkovsek G, Rubinic J (2003) Applicability
of the Gavrilovic method in erosion calculation using spatial data
manipulation techniques. In: De Boer D, Froehlich W, Mizuyama T,
Pietroniro A (eds) Erosion prediction in ungauged basins: integrat-
ing methods and techniques, vol 279. IAHS Publication, Walling-
ford, pp 22–233
Milanesi L, Clerici A, Pilotti M (2013) Il metodo dell’erosione
potenziale: proposte innovative ed applicazione in ambito alpino.
L’Acqua 1:37–47
View publication stats
L. Milanesi et al.
Pilotti M, Bacchi B (1997) Distributed evaluation of the contribution of
soil erosion to the sediment yield from a watershed. Earth Surf Proc
Land 22:1239–1251
Pilotti M, Gandolfi C, Bischetti GB (1996) Identification and analysis
of natural channel networks from digital elevation models. Earth
Surf Proc Land 21:1007–1020
Ranzi R, Le Hung T, Rulli MC (2012) A RUSLE approach to model
suspended sediment load into the Lo river (Vietnam): effects of
reservoirs and land use changes. J Hydrol 422–423:17–29
Tazioli A (2009) Evaluation of erosion in equipped basins,
preliminary results of a comparison between the Gavrilovic model
and direct measurements of sediment transport. Environ Geol
56:825–831
Zemljic M (1971) Calcul du debit solide - Evaluation de la vegetation
comme un des facteurs antierosifs. In: International Symposium
Interpraevent, Villach, Austria