© Acta hydrotechnica 21/34 (2003), Ljubljana
ISSN 1581-0267
37
UDK / UDC: 551.311.2:556.04:556.51:556.535(28.2.249)(497.14) Prejeto/Received: 30. 10. 2003
Predhodna objava – Preliminary scientific paper Sprejeto/Accepted: 17. 1. 2004
MERITVE EROZIJSKIH PROCESOV V EKSPERIMENTALNEM
POVODJU DRAGONJE, JZ SLOVENIJA
MEASUREMENTS OF EROSION PROCESSES IN THE EXPERIMENTAL
CATCHMENT OF THE DRAGONJA RIVER, SW SLOVENIA
Gregor PETKOVŠEK, Matjaž MIKOŠ
Prispevek obravnava nekatere rezultate meritev v povodju Rokave, ki je del eksperimentalnega
povodja Dragonje. Ve činoma so meritve potekale v času od jeseni 2000 do spomladi 2002. Merili
smo pretok vode, lebde čih in rinjenih plavin, dinamiko erozije sedimentov na klifih ter padavine, na
podlagi katerih smo dolo čili dejavnik erozivnosti padavin in odtoka R. Rezultati kažejo, da srednje
letno odplavljanje lebde čih plavin znaša okrog 1500 ton oziroma 750 kg/ha/leto. Pri meritvah klifov
smo uporabili kombinacijo fotogrametri čnih snemanj in meritev z erozijskimi žeblji či, kar nam je
dalo vpogled tako v skupno koli čino sproš čenega materiala kot v kratkotrajno dinamiko erozije s
klifov. Klifi so najaktivnejši pozno poleti, letno pa prispevajo v re čno mrežo okrog 100 ton
drobnega materiala letno. Erozivnost padavin je tako prostorsko kot časovno izrazito
neenakomerna. Najbolj erozivni so meseci od julija do novembra.
Klju čne besede: eksperimentalno povodje, Dragonja, Rokava, sedimenti, klifi, erozivnost padavin
The paper presents and analyses the results of some of the measurements in the Rokava catchment,
which is a part of the Dragonja catchment. Measurements were carried out mostly between autumn
2000 and spring 2002. The measured quantities included water discharge, suspended and bedload
discharge, dynamics of sediment production from cliffs, and measurements of rainfall, which were
used to compute rainfall and the runoff erosivity factor R. The results show that the mean annual
suspended sediment yield is around 1500 tonnes, which is equal to 750 kg/ha/year. Cliffs’ activity
was monitored with a combination of photogrammetric surveys and erosion pins, which gave us
both the total sediment production from cliffs and its short-term dynamics. Cliffs are most active in
late summer. They contribute approximately 100 tonnes of fine sediment into the stream network.
Rainfall erosivity is highly variable, both spatially and temporally. The rainfall erosivity is highest
in the period between July and November.
Key words: experimental catchment, Dragonja, Rokava, sediments, cliffs, rainfall erosivity
1. UVOD
Pri preoblikovanju zemeljskega površja ima
pomembno vlogo erozija. Erozija je posledica
razli čnih eksogenih dejavnikov, kot so na
primer delovanje teko če vode, snega, vetra,
temperaturnih nihanj in težnosti. Glede na
vzrok nastanka delita Pintar & Mikoš (1983)
erozijske pojave takole:
− pojavi kemi čnega, biološkega in
fizikalnega preperevanja,
− vetrna erozija,
− snežna erozija,
1. INTRODUCTION
Erosion of Earth’s surface plays an
important role in geomorphologic processes.
In general, erosion can be caused by different
exogenic factors, such as force of running
water, snow, wind, temperature oscillations
and gravity. Depending on the driving force,
Pintar & Mikoš (1983) consider the following
erosion processes:
−chemical, biological and physical
weathering,
− wind erosion,

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
38
− ledeniška erozija,
− vodna erozija,
− plazna erozija in
− podorna erozija.
V hidrologiji uporabljamo pojem erozija tal
tudi v ožjem pomenu, kot vodno erozijo. Z
njim ozna čujemo pojave površinskega spiranja
in odplavljanja zemljin zaradi delovanja
padavin in teko če vode. Erozija tal je v ve čini
primerov rezultat naravnih dejavnikov. V
nekaterih primerih pa jo pospešuje tudi
človekova dejavnost. Najbolj zna čilni primeri
so kmetijstvo, rudarstvo in gradbeništvo (Hahn
et al., 1994). Ta prispevek obravnava
predvsem procese vodne erozije, ki so na
obravnavanem obmo čju najbolj prisotni.
Dinamiko erodiranja zemljin sestavljajo
procesi sproš čanja, premeš čanja in odlaganja
(Meyer & Wischmeier, 1969). Pri sproš čanju
se predvsem zaradi dežnih kapelj delci lo čijo
od mati čnih tal. Površinski vodni tok zrna
premeš ča navzdol po pobo čju. Premestitvena
zmogljivost vodnega toka naraš ča s hitrostjo,
ko pa se hitrost zmanjša, obi čajno zaradi
zmanjšanja naklona pobo čja, nastopi odlaganje
(Hahn et al., 1994). Dinamiko in bilanco
sedimentov v povodju podaja slika 1.
Podrobnejši pregled procesov je podan v
Petkovšek (2000).
V tem prispevku podajamo rezultate
meritev odtoka vode, odplavljanja zemljin in
erozije s klifov. Meritve smo opravili na
severovzhodnem delu eksperimentalnega
povodja Dragonje, to je v povodju Rokave
(slika 2). Povodje Rokave je dobro raz členjen
gri čevnat svet, s površino 20,4 km
2
 in srednjo
nadmorsko višino 250 m. Najnižja to čka je
soto čje z Dragonjo (75 m NMV), najvišja pa
415 m NMV. Povpre čni vzdolžni padec doline
je v zgornjem delu 2,2 %, v spodnjem delu pa
1,4 %. Srednji nagib površja je 25 %. Meritve
smo izvajali v sodelovanju z Vrije Universiteit
Amsterdam v okviru projekta "Dragonja:
Forest – Soil – Water – Climate Interactions"
(Dragonja: medsebojni vplivi gozda, tal, vode
in klime).
Povodje Dragonje je v zadnjih desetletjih
doživelo ve čje spremembe rabe tal, kar je
vplivalo na režim odtoka vode ter sproš čanje
− snow erosion,
− glacial erosion,
− fluvial erosion,
− landslides and rockfalls.
In hydrology, the term soil erosion is
usually used in a narrower context, as water
erosion. It denotes the processes of superficial
soil detachment and transport due to raindrop
impact and running water. Soil erosion is a
natural process. In some cases, however,
human activity accelerates the natural
phenomenon. These activities are agriculture,
mining and construction (Hahn et al., 1994).
This paper analyses the processes of water
erosion, which are most important in the
studied area.
The soil erosion process consists of soil
detachment, transport and deposition (Meyer
& Wischmeier, 1969). Detachment is a process
caused mainly b y raindrop impact, which
separates the soil particles from the ground.
Overland flow transports the particles
downslope. The transport capacity of the
overland flow increases with flow velocity.
Downslope, the angle of the slope usually
decreases, therefore flow velocity also
decreases and deposition takes place (Hahn et
al., 1994). The sediment dynamics in a
catchment is shown in Figure 1. More details
are given in Petkovšek (2000).
In this paper, the results of the
measurements of water runoff, sediment yield
and sediment production from cliffs are given
and discussed. Measurements took place in the
Rokava catchment, which is a part of the
Dragonja experimental catchment (see Figure
2). The Rokava catchment is formed of hills
with a well-developed stream network. The
area of the catchment is 20.4 km
2
, the mean
elevation is 250 m a.s.l. The lowest point is the
confluence with the Dragonja (75 m a.s.l.), the
highest point is at 415 m a.s.l. The average
longitudinal slope of the valley is 2.2 % in the
upper part and 1.4% in the lower part. The
average slope of the surface is 25 %. The
measurement campaign was a joint project
with the Vrije Universiteit Amsterdam called
"Dragonja: Forest – Soil – Water – Climate
Interactions".
In the last decades, the Dragonja catchment
underwent significant changes in land use.
This also influenced water runoff and sediment

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
39
in odplavljanje zemljin (Globevnik & Sovinc,
1998). Viri navajajo na podlagi vrednotenja z
Gavrilovi ćevo ena čbo (Gavrilovi ć, 1970), da
se je sproš čanje od sedemdesetih let (PUH,
1971) do danes zmanjšalo za skoraj 70 %
(Globevnik, 2001).
yield regime (Globevnik & Sovinc, 1998).
Authors that used the Gavrilovi ć e quation
(Gavrilovi ć, 1970) state that sediment
production (PUH, 1971) for nearly 70 %
decreased from the seventies (Globevnik,
2001).
STRUGA,
POPLAVNO OBMO ČJE
IZTOK IZ POVODJA
sproš čanje
odplavljanje
POBO ČJE
Slika 1. Dinamika, bilanca ter obmo čja odlaganja in erodiranja sedimentov v povodju. Puš čice,
obrnjene navzdol, ponazarjajo premeš čanje, puš čice, obrnjene navzgor, pa odlaganje. Debelina
puš čic je v razmerju s koli čino sedimentov za namišljen primer. Prirejeno po Reid & Dunne (1996).
CHANNEL
OVERBANK AREA
OUTFLOW FROM CATCHMENT
production
yield
SLOPE
Figure 1. Schematics of sediment dynamics in a catchment. Arrows directed downwards represent
erosion; arrows directed upwards represent deposition. The thickness of the arrows represents the
approximate magnitude for a typical catchment. After Dunne & Reid ( 1996).

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
40
Slika 2. Položaj povodja Rokave znotraj eksperimentalnega povodja Dragonje
v Slovenski Istri brez dela povodja na Hrvaškem.
Figure 2. Location of the Rokava catchment inside the experimental watershed of the Dragonja
in Slovene Istria without the part in Croatia.
Slika 3. Merske lokacije.
Figure 3. Locations of measurements.

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
41
2. METODE
Meritve, s katerimi smo ocenjevali
sproš čanje in odplavljanje zemljin, so
vklju čevale meritve padavin, meritve pretoka
vode in lebde čih plavin ter meritve erozije s
klifov. V podpovodju Rokave smo izbrali tri
klife z razli čno stopnjo aktivnosti. Meritve so
pokrivale obdobje dveh let med letoma 2000
in 2002. Merske lokacije so podane na sliki 3.
2.1 MERITVE PRETOKA, LEBDE ČIH IN
RINJENIH PLAVIN
Na lokaciji Rokava je bila postavljena
vodomerna postaja z avtomatskim
vzor čevalnikom ISCO 3700. V stabilnem
reguliranem delu struge, širine okrog 5 m, je
bil zgrajen široki prag, ki ga je za potrebe
projekta zgradilo lokalno vodnogospodarsko
podjetje Hidro Koper. Za njim sta bili
postavljeni dve tla čni sondi Druck Ltd. PDCR-
830 (Hobby & Minneboo, 2001). Prva sonda
beleži gladino vode v konstantnih časovnih
intervalih (10 oziroma 20 minut). Druga sonda
beleži gladino vode v daljših časovnih
intervalih (1 ura), vendar tudi vsaki č, ko je
sprememba ve čja od predpisane vrednosti (0,5
cm). Podatki z druge sonde se shranjujejo v
pomnilnik instrumenta Campbell 21X, ki
obenem ob vsaki spremembi gladine za ve č
kot 5 cm sproži proceduro odvzema vzorca
vode z avtomatskim vzor čevalnikom vode
ISCO 3700. Ta lahko brez praznjenja odvzame
najve č 24 vzorcev po 500 ml, kar zagotavlja
precejšnjo avtonomnost vzor čevanja lebde čih
plavin. Zajemna glava je bila postavljena pod
širokim pragom, kjer je voda dobro
premešana. S tem smo zagotovili ustrezno
reprezentativnost vzorca. Velikost odvzetih
vzorcev je bila med 400 in 500 ml.
Koncentracija C
s
 [g/l] lebde čih plavin je bila
dolo čena tako, da je bila najprej izmerjena
prostornina vzorca V
s
, nato pa vzorec izparjen
pri 105 °C, sušina pa stehtana. Koncentracija
je bila izra čunana kot razmerje med maso
sušine m
s
 in prostornino vzorca V
s
.
Pretoki vode so bili dolo čeni s pomo čjo
preto čne krivulje, ki je bila dolo čena z
2. METHODS
The aim of the measurements was to assess
the sediment production and sediment yield
from the catchment. Measurements included
measurements of rainfall, water and suspended
sediment discharge and monitoring o f
sediment production from cliffs in the period
2000–2002. Three cliffs with different
activities were chosen in the Rokava sub-
catchment. Locations are shown in Figure 3.
2.1 MEASUREMENTS OF WATER
DISCHARGE, SUSPENDED SEDIMENT
AND BEDLOAD
At location Rokava, a water level recorder
and automatic water sampler ISCO 3700 were
set up. In an approximately 5 m wide, stable
regulated part of the stream, a broad crested
weir was constructed by local water
management company Hidro Koper. Two
pressure transducers of type Druck Ltd.
PDCR-830 were placed just upstream (Hobby
& Minneboo, 2001). The first transducer
recorded the water level at constant intervals
(10 or 20 minutes). The second transducer
recorded the water level every hour, but also at
every 0.5 cm change in water level. The data
from the second transducer was logged to the
Campbell 212X datalogger. The datalogger
triggered the water sampler every time when
the change in water level exceeded 5 cm. The
automatic water sampler ISCO 3700 could
take 24 samples of 500 ml without being
emptied. This allowed for satisfactory
automatisation of the process. The intake was
placed just under the broad crested weir, where
the water was well mixed. Thus the
representativeness of the samples was ensured.
The volume of samples was between 400 and
500 ml. Concentration C
s
 [g/l] of suspended
sediment was determined by determining the
exact volume of the sample V
s
. Then, the
sample was evaporated at 105°C and the dry
residual weighted (m
s
). The concentration was
then calculated as the ratio between m
s
 and V
s
.
The discharge of water was determined
from the discharge curve. The discharge curve
was determined by occasional measurements

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
42
občasnimi meritvami pretokov. Pretoki so bili
izra čunani s trapezno integracijo specifi čnih
pretokov po širini, ti pa so bili izra čunani iz
izmerjene hitrosti in globine. Hitrosti so bile
merjene z elektromagnetnim merilcem hitrosti
OTT Nautilus C2000 (merilno obmo čje 0 ÷
2,5 m/s).
Dolo čitev prostornine odtoka za posamezen
poplavni val je potekala po naslednjem
postopku. Najprej sta bila dolo čena za četni t
z
in končni čas t
k
 poplavnega vala. Za za četek
poplavnega vala je veljal tisti čas pred
nastopom konice, ko je bil pretok najmanjši.
Za čas upadanja t
r
 (od vrha do konca
poplavnega vala) je bila izbrana fiksna
vrednost, ki je bila na podlagi primerjav
poplavnih valov ocenjena na 12 ur. Nadalje je
bilo predpostavljeno, da je bazi čni odtok Q
b
 do
nastopa vrha poplavnega vala enak pretoku ob
za četku poplavnega vala, od tam naprej pa
linearno naraš ča do vrednosti na koncu
poplavnega vala. Kon čno je bila prostornina
poplavnega vala izra čunana z integracijo
razlike med izmerjenim pretokom v strugi Q in
bazi čnim odtokom Q
b
:
of discharge at different water levels. The
discharges were computed from the measured
velocities and cross section geometry.
Velocities were measured by electromagnetic
flow sensor OTT Nautilus C2000 (range 0 ÷
2.5 m/s).
The volume of surface runoff for each flood
wave was computed by the following
procedure. First, the time of the beginning t
z
and the time of end t
k
 of the flood wave was
determined. The time of beginning was
defined as the time before the peak with lowest
discharge. For the recession time t
r
 (from time
of peak to ending time) a fixed value was
chosen. The t
r
 was estimated to 12 hours.
Further, we assumed that baseflow Q
b
 in the
time period between the beginning and peak
was equal to the discharge at the beginning t
z
of the flood wave. From then on, it increased
linearly till the end of the flood wave. The
volume of surface runoff was computed by
integrating the difference between the total
discharge Q and baseflow Q
b
:
∫
⋅ − =
k
z
t
t
b
dt Q Q V )( (1)
Letno stopnjo premeš čanja rinjenih plavin
smo ocenili na podlagi zaprojevanja
zaplavnega prostora za pragom vodomerne
postaje.
2.2 MERITVE SPROŠ ČANJA S KLIFOV
Pri meritvah klifov smo uporabili zanimivo
in uspešno kombinacijo fotogrametri čne
meritve in meritev z erozijskimi žeblji či. Tako
smo izmerili tako skupno sproš čanje kot tudi
kratkotrajno dinamiko sproš čanja sedimenta s
klifa.
V povodju Rokave smo na podlagi kart
identificirali aktivne klife, to je take, ki niso v
zaraš čanju in lahko na njih še vedno
pri čakujemo velike stopnje sproš čanja. Vsem
klifom smo dolo čili velikost, na najve čjem
klifu (Rokava-1, sliki 3 in 4) pa smo opravili
tudi podrobno fotogrametri čno meritev v treh
časovnih presekih.
The annual bed load rate was estimated
from the volume of the sediment deposited
behind the broad crested weir.
2.2 SEDIMENT PRODUCTION FROM
CLIFFS
For the monitoring of cliffs an interesting
and powerful combination of photogrammetric
surveys and measurements with erosion pins
was used. In this way, we were able to
measure both total sediment production as well
as its short-term dynamics.
The active cliffs were identified from maps.
Every bare steep surface that was not
vegetated in the lower part was considered an
active cliff. The areas of all active cliffs were
measured. On the biggest cliff (Rokava-1,
Figures 3 and 4) three photogrammetric
surveys were performed.

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
43
Slika 4. Klif Rokava-1 v pogledu. Legenda: E – erodirajo či deli klifa; N – odlaganje sedimentov.
Šrafirano obmo čje smo spremljali z erozijskimi žeblji či. Koordinate so lokalne.
Figure 4. Front view of the cliff Rokava- 1. Legend: E – eroding areas of the cliff,; N – deposition
areas. Hatched area was monitored by erosion pins. Coordinates are local.
Na manjših klifih smo izmerili nekaj
zna čilnih to čk, s čimer smo lahko ocenili
njihovo velikost. Za dolo čanje razdalj smo
uporabili laserski razdaljemer DISTO pro
(Leica Geosystems, 2001). Natan čnost
razdaljemera, kot navaja proizvajalec, je ± 5
mm, merimo pa lahko razdalje do 100 m.
Višino to čk smo dolo čili po ena čbi (2).
Oznake razdalj so prikazane na sliki 5. Kljub
veliki natan čnosti razdaljemera, zaradi manj
zanesljivih meritev višin ocenjujemo
natan čnost metode na ± 10 cm.
The area of smaller cliffs was determined
by measuring a number of topographically
characteristic points. The distances were
measured by the laser distance meter DISTO
pro (Leica Geosystems, 2001). The accuracy
of the device, as stated by the manufacturer, is
± 5 mm. The distances up to 100 m can be
measured. The elevation of points was
determined by equation (2) and as shown in
Figure 5. In spite of the very accurate distance
meter, the overall accuracy was estimated to ±
10 cm due to less accurate measurement of
elevation.
1
1
0 1
) (
b
d
h h z ⋅ − = (2)
Meritve najve čjega klifa ob Rokavi smo
naro čili pri Geodetskem inštitutu Slovenije
(GI, 2001; 2002a; 2002b). Natan čnost
dobljenih 3D-to čk je v zgornjem delu do ± 10
cm, v spodnjem delu pa ± 2 cm.
Na podlagi dobljenih to čk smo izdelali 3D-
model tipa nepravilne trikotniške mreže (TIN).
Dolo čili smo nagib ter površino klifa v tlorisu
in pogledu.
Pri klifu Rokava-1, ki smo ga posneli v treh
časovnih presekih, smo analizirali tudi
Photogrammetric survey of the biggest cliff
Rokava-1 was carried out by the Geodetic
institute of Slovenia (GI, 2001; 2002a; 2002b).
The accuracy of the obtained 3D points was ±
10 cm in the upper part and ± 2 cm in the
lower part.
From the measured points a 3D model of
cliff surface was constructed as triangulated
irregular network (TIN). Areas in plan and
front view were calculated.
The change surface and volume was

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
44
spremembo površine in prostornine. Erozijo in
odlaganje na tem klifu smo spremljali tudi s
pomo čjo erozijskih žeblji čev. Postavili smo 28
žeblji čev v žlebi čih ter v medžlebi čnem
prostoru, in sicer tako na obmo čju, ki je bilo
deloma zaš čiteno z vegetacijo (šopi trave), kot
na nezaš čitenem obmo čju. Žeblji če smo
razmestili v spodnjem delu klifa na njegovem
skrajnem koncu v soto čni smeri (slika 4).
analysed for the cliff Rokava-1, where three
time records of the cliff were available. Also,
short-term erosion or sedimentation was
monitored using erosion pins. A network of 28
pins was set up in three different areas of the
cliff: in rills, in the bare interrill area and in the
vegetated interrill area. Pins were placed in the
lower part of the cliff at its downstream most
corner (Figure 4).
h
1
h
0
z
b
1
d
1
Slika 5. Dolo čitev višine merjene to čke z s pomo čjo razdaljemera.
Figure 5. Determination of elevation by distance meter.
2.3 MERITVE EROZIVNOSTI PADAVIN
Padavine smo merili z avtomatskimi
dežemeri Siap WL-8100 z lo čljivostjo 0,1 mm
dežja. Dežemeri so bili postavljeni na izbranih
lokacijah v Kocjan či čih, Marezigah, ob
Rokavi, ob Dragonji in na Koštaboni. Lokacije
so podane na sliki 3.
Erozivnost smo izrazili z dejavnikom R
(Renard et al., 1997). Padavinski podatki iz
dežemerov so bili najprej razdeljeni na 15-
minutne intervale. Nato so bili iz njih izlo čeni
erozivni dogodki oziroma nalivi. Po Renard et
al. (1997) je erozivni dogodek definiran kot
dogodek, v katerem pade skupaj najmanj 12
mm padavin oziroma najve čja skupna koli čina
padavin v pol ure preseže 6 mm. Dva dogodka
sta lo čena, če med njima obstaja šesturni
interval, v katerem pade manj kot 1,2 mm
padavin.
2.3 RAINFALL EROSIVITY
Rainfall was measured by automatic
raingauges Siap WL-8100 with a resolution of
0.1 mm. Raingauges were placed in selected
locations in Kocjan či či, Marezige, at Rokava,
Dragonja and Koštabona. Locations are shown
in Figure 3.
Rainfall erosivity was estimated with the R
factor (Renard et al., 1997). Raw rainfall data
was distributed in 15 minute time intervals.
Then the erosive events were determined.
Following Renard et al. (1997), an erosive
event is defined as event, where total amount
of precipitation is more than 12 mm or when
the maximum amount of precipitation in 30
minutes exceeds 6 mm. Two events are
separated if a six-hour time interval exists
between them with less than 1.2 mm of
precipitation.

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
45
Dejavnik R [MJ/ha mm/h] je za en naliv
definiran kot produkt energije padavin E
[MJ/ha] in najve čje 30-minutne intenzitete
naliva I
30
 [mm/h]:
R factor [MJ/ha mm/h] for a single event is
defined as a product of rainfall energy E
[MJ/ha] and maximum 30 minute intensity I
30
[mm/h]:
30
I E R ⋅ = (3)
Energija padavin E je odvisna od trenutne
intenzitete naliva i* [mm/h]. Enotsko energijo
padavin e [MJ/ha·mm] na enoto površine (ha)
in enoto padavin (mm) dolo čimo po empiri čni
ena čbi:
The rainfall energy is a function of
instantaneous intensity i* [mm/h]. Unit rainfall
energy e [MJ/ha·mm] per unit of area (ha) and
unit of rainfall (mm) is calculated by the
following equation:
*)) 05 . 0 exp( 72 . 0 1 ( 29 . 0 i e ⋅ − ⋅ − ⋅ = (4)
Letna erozivnost je seštevek erozivnosti
vseh posameznih nalivov v izbranem letu.
3. REZULTATI IN RAZPRA VA
3.1 MERITVE PRETOKOV VODE,
RINJENIH IN LEBDE ČIH PLAVIN
Struga Rokave je del leta suha. Skupen čas,
ko je v sezoni 2000/01 po strugi tekla voda, je
bil 248 dni. Srednji letni pretok, če ra čunamo
samo obdobje, ko je bila v strugi voda, je 428
l/s. Če to prera čunamo na celo leto (365 dni),
dobimo srednji letni pretok 291 l/s. Najve čji
zabeleženi pretok 19,7 m
3
/s v omenjenem
obdobju je nastopil 4. 11. 2000. Srednja letna
visoka voda v merilnem prerezu znaša Q
2 
=
16.3 m
3
/s, visoka voda s petletno povratno
dobo pa Q
5 
= 25 m
3
/s (Petkovšek, 2002).
Pri šestih poplavnih valovih v tem obdobju
je bila merjena tudi koncentracija lebde čih
plavin in izra čunan njihov masni pretok
(kalnost). Vrednosti so podane v preglednici 1.
Pri nekaterih dogodkih so bile izmerjene
koncentracije nekajkrat ve čje od povpre čja.
Pri poplavnem valu 14. 9. 2001 je v
naraš čajo či veji, pri pretoku približno 10 m
3
/s,
vodomerna postaja prenehala pravilno
delovati. Zato podatki o najve čjem pretoku
temeljijo na oceni najvišje dosežene kote vode
(sledi na obrežnem grmovju), prostornina
poplavnega vala pa je bila ocenjena na podlagi
trajanja poplavnega vala in ocene koeficienta
odtoka.
Annual erosivity is the sum of erosivities of
all events in the selected year.
3. RESULTS AND DISCUSSION
3.1 WATER DISCHARGE, SUSPENDED
SEDIMENT AND BEDLOAD
During summer, there is no water flow in
the Rokava stream. Total time with water flow
in the season 2000/01 was 248 days with a
mean discharge of 428 l/s. If this is averaged
over the whole year (365 days), the mean
value is 291 l/s. The maximum recorded
discharge of 19.7 m
3
/s in the period occurred
on November 4, 2000. Discharges with
characteristic return periods were estimated by
Petkovšek (2002). The discharge with a return
period of two years equals 16.3 m
3
/s, and that
of five years equals Q
5 
= 25 m
3
/s.
Concentration of suspended sediment and
their discharges was measured for six events.
Values are given in Table 1. At some events,
the measured concentrations exceeded the
average value by several times.
For the flood wave on September 14, 2001
the water level recorder stopped functioning
properly in the rising limb at the discharge of
approximately 10 m
3
/s. Therefore, the data on
peak discharge is based on the highest water
level reached, which in turn was estimated by
traces left on riparian vegetation. The runoff
volume was estimated on the base of the flood
wave duration and an estimate of the runoff
coefficient.

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
46
Preglednica 1. Pregled poplavnih valov na vodomerni postaji Rokava, pri katerih so bile vzor čevane
lebde če plavine. Oznake: izmerjene koncentracije: C
m
 – srednja vrednost, C
max
 – najve čja vrednost;
M
SS
 – skupna masa premeš čenih lebde čih plavin, Q
max
 – najve čji pretok vode, V
q
 – prostornina
površinskega odtoka, P – višina padavin.
Table 1. Overview of sampled flood events at Rokava sampling station. Explanation of symbols:
concentrations: C
m
 – mean, C
max
 – maximum; M
SS
 – total mass of transported suspended
sediment, Q
max
 – peak flow, V
q
 – volume of surface runoff, P – precipitation.
datum / date C
m
C
max
M
SS
Q
max
V
q
P
d.m.yy kg/m
3
kg/m
3
1000 kg m
3
/s 1000 m
3
mm mm
10.10.00 0.72 2.72 61.3 5.6 92 4.7 51
4.11.00 2.72 3.33 759 19.7 275 14.0 79
25.11.00 1.04 5.98 96.5 5.0 89 4.5 29
14.9.01 4.73 14.91 1800* 20* 380* 19.5* 108
7.2.02 0.53 0.88 32.5 1.77 61.8 3.0 23
9.4.02 0.06 0.07 4.2 1.65 66.1 3.2 48
Na podlagi prikazanih rezultatov v
preglednici 1 lahko zaklju čimo, da manjši
poplavni valovi odplavijo do 100 ton lebde čih
plavin, ve čji pa tudi 1000 in ve č ton. V obeh
opazovanih sezonah se je pojavil en velik
poplavni val in nekaj (5–10) manjših
poplavnih valov, ki premeš čajo lebde če
plavine. Tako lahko letno odplavljanje
lebde čih plavin v opazovanem obdobju s
povodja Rokave ocenimo na okrog 1500 ton
letno, ta vrednost pa je najbolj odvisna od
najve čjega dogodka v letu. Leto 2001 je bilo,
glede na padavine, sicer nekoliko
podpovpre čno (na lokaciji klimatološke
postaje Portorož – Letališ če 978 mm v
primerjavi z dolgoletnim povpre čjem 1013
mm). Meseci oktober, november in december
2000 pa so bili nadpovpre čno mokri (na
lokaciji Portorož – Letališ če 652 mm v
primerjavi z dolgoletnim povpre čjem v teh
mesecih 318 mm).
Razmerje med koncentracijo lebde čih
plavin C in pretokom Q niha, zlasti med
razli čnimi poplavnimi valovi. Tako je bila pri
dogodku dne 10. 10. 00 koncentracija lebde čih
plavin pri pretoku Q = 3 m
3
/s le okrog 0,5 g/l,
medtem ko je bila pri dogodkih 4. 11. 2000 in
14. 9. 2001 okrog 2 g/l. Opazimo lahko tudi,
da se koncentracija lebde čih plavin s pretokom
včasih ne zvišuje, najvišje koncentracije se
pojavljajo pri relativno nizkih pretokih. To
From Table 1 it can be concluded that the
suspended sediment discharge at smaller
events is approximately 100 tonnes, while at
bigger events it is above 1000 tonnes. In both
monitored seasons one big event and several
(5–10) small events with ability to transport
suspended sediment occurred. Therefore, the
suspended sediment yield in the monitored
period can be estimated to 1500 tonnes per
year. Variation mostly depends on the biggest
event in a year. The precipitation in the year
2001 was a bit below the mean value. At the
location of the climatological station in
Portorož Airport, the annual precipitation was
978 mm compared to the long-term average of
1013 mm. The months of October, November
and December 2000 were above the long-term
average (at Portorož Airport, 652 mm
compared to the long-term average in the same
months of 318 mm).
Relation between the concentration of
suspended sediment C and the discharge Q
varies between events. The concentration at
the discharge Q = 3 m
3
/s was around 0.5 g/l
for the event of October 10, 2000, while it was
around 2 g/l for events of November 4, 2000
and September 14, 2001. Often the
concentration was not increasing with the
discharge and the peak in concentration
occurred at relatively low discharges. This is
especially true for the event of November 4,

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
47
velja npr. za dogodek 4. 11. 2000. Izmerjeni
podatki torej kažejo, da omejitveni dejavnik
premeš čanja lebde čih plavin ni premestitvena
zmogljivost, temve č dotok sedimentov. To
potrjuje tudi dejstvo, da na dnu skoraj ni
mogo če najti drobnih zrn plavin. Ravno tako
lahko sklepamo, da erozija brežin, ki je
odvisna predvsem od pretoka vode, ne
prispeva bistvenega deleža lebde čih plavin.
Zaplavni prostor za pragom na lokaciji
vodomerne postaje Rokava se je zapolnil v
času dogodka 4. 11. 2000 (vrh poplavnega
vala Q
max 
= 19,7 m
3
/s). Prostornino plavin za
pragom smo ocenili na 50 m
3
. Približno enaka
koli čina rinjenih plavin se je odložila na
položnem delu med pragom in stopnjo, ki leži
približno 50 m dolvodno. V času tega
poplavnega vala se je po strugi skozi presek tik
nad pragom torej premestilo najmanj 100 m
3
rinjenih plavin.
2000. The measured data show that the
sediment discharge is not transport capacity
limited, but supply limited. This can also be
confirmed by the fact that there is very little
fine sediment in the streambed. It can also be
concluded that the bank erosion, which is
mostly water discharge dependent, does not
contribute a considerable amount of sediment.
The deposition volume behind the broad
crested weir at the site of Rokava water level
recorder was filled during the event of
November 4, 2000 (peak flow Q
max
= 19.7
m
3
/s). Volume of deposited bedload was
estimated at 50 m
3
. Approximately the same
volume was deposited between the broad
crested weir and a step located approximately
50 m downstream. Therefore, we estimate that
during that event at least 100 m
3
 of bedload
was transported.
Slika 6. Deli klifa glede na dinamiko sedimentov.
Figure 6. Parts of cliff regarding sediment dynamics.
POBO ČJE
SLOPE
zgornje obmo čje
upper area
spodnje obmo čje
 lower area
STRUGA
CHANNEL
KLIF
CLIFF

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
48
Preglednica 2. Dimenzije in nagib klifa Rokava-1.
Table 2. Dimensions and slope of the Rokava- 1 cliff.
zgornje obmo čje (erozija)
upper area (erosion)
spodnje obmo čje (odlaganje)
lower area (deposition)
skupaj
total
površina – tloris A
z
area – plan view A
z
2682 m
2
878 m
2
3560 m
2
površina – pogled A
y
area – front view A
y
3685 m
2
745 m
2
4430 m
2
nagib
slope
1.48 0.87 1.32
3.2 EROZIJA S KLIFOV
Na povodju Rokave smo identificirali štiri
erozijsko aktivne klife: Rokava-1, Rokava-2,
Rokava-3 in Lopar (slika 3). Klifi so
sestavljeni iz dveh obmo čij (slika 6): zgornje
obmo čje se ve činoma erodira, na spodnjem
obmo čju klifa se sedimenti za časno odlagajo.
Na podlagi terenskih opazovanj smo prišli
do naslednjih ugotovitev o dinamiki
sedimentov na klifih:
1. Na zgornjem delu klifa prihaja do
erozije, predvsem zaradi preperevanja
in zdrsov manjših zemljinskih mas.
2. Drobnejše frakcije se odložijo na
spodnjem delu klifa, grobe frakcije pa
zaradi ve čje vztrajnosti padejo
neposredno v strugo.
3. Iz spodnjega dela sedimente v strugo
odnaša predvsem površinska vodna
erozija (medžlebi čna in žlebi čna
erozija).
3.2.1 KLIF ROKAVA-1
Klif Rokava-1 je najve čji. Podatki o tem
klifu so bili dobljeni s fotogrametri čnim
snemanjem. Klif je visok do 50 m, širina pa je
180 m. Pogled z zna čilnimi deli klifa je
prikazan na sliki 4. Spodnji del klifa v
protito čni smeri je precej zaraš čen, tako da tam
meritve nismo opravili. Preglednica 2 podaja
nekatere zna čilnosti klifa.
Klif je spodaj zaraš čen do x = 980 m
(lokalne koordinate, obmo čje 1E na sliki 4),
zato lahko sklepamo, da je koli čina
erodiranega materiala, ki doseže strugo, na tem
3.2 EROSION FROM CLIFFS
Four active cliffs were identified in the
Rokava catchment: Rokava-1, Rokava-2,
Roaka-3 and Lopar (Figure 3). Cliffs consist of
two areas (Figure 6): the upper area, where the
area is mainly eroded, and the lower area,
where sediment is temporarily deposited.
Field observations helped us reach the
following conclusion about sediment dynamics
on cliffs:
1. In the upper part, the sediment is mostly
eroded due to decay and small
landslides.
2. Fine sediment is deposited in the lower
part of the cliff, while bigger grains tend
to fall directly into the streambed.
3. From the lower part, sediment is eroded
into the stream mostly by soil erosion
processes caused by rainfall (rill and
inter-rill erosion).
3.2.1 CLIFF ROKAVA-1
Rokava-1 cliff is the biggest cliff in the
catchment. Topography of the cliff was
obtained by a photogrammetric survey. The
cliff is up to 50 m high and 180 m wide. The
front view with indicated characteristic parts is
shown in Figure 4. In the upstream direction,
the lower part is considerably vegetated and no
survey of that part was performed. Table 2
gives some data about the cliff.
The cliff is vegetated up to x = 980 m (local
coordinates, area 1E in Figure 4). It can be
therefore assumed that the amount of sediment
reaching the stream in this part is negligible.

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
49
delu majhna. Med x = 980 m in x = 1000 m je
na višini z = 120 m klif precej položen in
zaraš čen, tako da se sedimenti, ki se morebiti
sprostijo iz višje leže čih delov, tam ustavijo in
prav tako ne dosežejo struge (obmo čje 3E na
sliki 4). Površine in nagibi posameznih delov
klifa so podani v preglednici 3.
At elevation z = 120 m between x = 980 m and
x = 1000 m the cliff surface is nearly flat. The
sediment that is eroded from higher parts is
deposited there without the possibility of being
flushed to the stream (area 3E in Figure 4).
Areas and slopes of individual parts of the cliff
are given in Table 3.
Preglednica 3. Površine in nagibi obmo čij klifa Rokava-1 (glej tudi sliko 4).
Table 3. Rokava- 1 cliff: areas and slopes (see also Figure 4).
obmo čje / area A
z
 [m
2
] A
y
 [m
2
]n a g i b I [-]
1E 1096 1593 1.56
2E 978 1222 1.35
2N 390 370 0.97
3E 194 271 1.50
4E 414 600 1.56
4N 488 373 0.78
Preglednica 4. Prostornine erodiranega in nanesenega materiala na klifu Rokava-1
med 27. 9. 2001 in 5. 2. 2002.
Table 4. The volumes of eroded and deposited material. Cliff Rokava- 1,
time span September 27, 200 1 to February 5, 2002.
V [m
3
]
erodirajo či del
eroding area
nasuti del
deposition area
skupaj
total
erodirani material
eroded material
–10 –54 –73
odloženi material
deposited material
3 175 178
skupaj
total
–7 112 105
Razlike med časovnima presekoma 27. 9.
2001 in 5. 2. 2002 so podane v preglednici 4.
Med tem časom ni bilo intenzivnih deževij in
visokovodnih valov, zato lahko sklepamo, da
so spremembe posledica preperevanja
materiala oziroma drugih oblik erozije (zdrsi
ipd.). Med časovnima presekoma 5. 2. 2002 in
27. 8. 2002 pa je zaradi površinske vodne
erozije na nekaterih nasutih delih klifa prišlo
tudi do odnašanja materiala.
Zaradi kratkih časovnih obdobij med
snemanji in slabše natan čnosti meritve v
zgornjem delu klifa so rezultati za erodirajo či
The differences between time sections
September 27, 2001 and February 5, 2002 are
given in Table 4. During that time, there were
no intensive rainfall and flood events.
Therefore, it can be concluded that the
differences are a consequence of material
decay and small landslides. On the other hand,
between February 5, 2002 and August 27,
2002 rainfall did cause some erosion from the
deposition parts as well.
As the time span between surveys is short
compared to the accuracy of the surveys in the
upper part of the cliff, the results for erosion in

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
50
del manj zanesljivi. Pri dolo čanju specifi čnega
letnega sproš čanja smo se zato oprli na
rezultate odlaganja na spodnjem delu klifa v
prvem polletnem obdobju (27. 9. 2001–5. 2.
2002), ko erozije na tem obmo čju klifa skoraj
ni bilo. Če predpostavimo, da se je ves
sproš čen material odložil na klifu, lahko
izra čunamo, da je specifi čno letno sproš čanje
na enoto tlorisne površine okrog 500 m
3
/ha
oziroma na enoto površine v pogledu okrog
400 m
3
/ha.
3.2.2 DRUGI KLIFI
Poleg velikega klifa smo izmerili tudi druge
manjše aktivne klife. Preglednica 6 podaja
nekatere dimenzije in nagibe klifov. Pri klifu
Rokava-3 obmo čja nasipanja nismo zaznali v
nobenem obdobju. Zato smo sklepali, da klif
ne prispeva pomembnejše koli čine sedimentov
v strugo.
Iz preglednice 6 vidimo, da so v primerjavi
s klifom Rokava-1 drugi klifi mnogo manjši.
Njihova skupna površina je približno petkrat
manjša. Iz nagibov spodnjega dela klifa lahko
izra čunamo strižni kot erodiranega materiala.
Ta znaša med 35° (Rokava-2) in 41° (Rokava-
1). To so precej visoke vrednosti, ki pa so
glede na obliko (ravnokar odlomljen oglat
material) pri čakovane.
3.2.3 PRIMERJAVA S PODOBNIMI
MERITVAMI V HRVAŠKI ISTRI
O meritvah sproš čanja s klifov poro čajo iz
hrvaškega dela Istre (Jurak et al., 2002), ki so
od leta 1995 s fotogrametri čno metodo merili
sproš čanje s klifa v Sv. Donatu (A
z
 = 2191 m
2
,
A
y
 = 3856 m
2
) in erozijske eksperimentalne
ploskve v Abramih (A
z
 = 8,9 m
2
, A
y
 = 3,7 m
2
).
Tam znaša specifi čno letno sproš čanje med
393 in 497 m
3
/ha v tlorisu oziroma med 200 in
269 m
3
/ha v pogledu. Vrednosti so podobne,
kot smo jih dobili na klifu Rokava-1. To je
pri čakovano, saj so klifi tako v topografskem
kot geološkem pogledu podobni.
the upper part are less reliable. Therefore, the
results from the lower part in first half-year
(27 Sept. 2001–5 Feb. 2002) were chosen for
calculation of specific (per year and area)
sediment production from the cliff. There was
hardly any erosion of sediment from this part.
If we assume that all the sediment eroded from
the upper part of the cliff was deposited in the
lower part, the specific annual sediment
production can be estimated at 500 m
3
/ha of
area in plan view or 400 m
3
/ha in front view.
3.2.2 OTHER CLIFFS
In addition to the biggest cliff, other cliffs
were also measured. Their sizes and slopes are
given in Table 6. Cliff Rokava-3 had no
deposition area and therefore it was concluded
that the amount of the sediment contributed to
the stream network is negligible.
From Table 6 it can be seen that the other
cliffs are much smaller in comparison to the
Rokava-1 cliff. Their total area is about five
times smaller. From the slope of the lower part
of a cliff angle of repose of the eroded material
can be calculated. It is between 35° (Rokava-
2) and 41° (Rokava-1). These values are high
but considering the shape (freshly broken
angular material) they are expected.
3.2.3 A COMPARISON TO SIMILAR
MEASUREMENTS IN CROATIAN ISTRIA
Jurak et al. (2002) report about sediment
production from cliffs in the Croatian Istria.
From 1995, photogrammetrical surveys of the
cliff in Sv. Donat (A
z
 = 2191 m
2
, A
y
 = 3856
m
2
) and experimental erosion plot in Abrami
(A
z
 = 8.9 m
2
, A
y
 = 3.7 m
2
) were performed.
Specific annual sediment production from
these cliffs is between 393 and 497 m
3
/ha in
plan view and between 200 and 269 m
3
/ha in
front view. These values are similar to the
ones obtained from Rokava-1 cliff. Similarity
could be expected since the cliffs are similar in
topography and geology.

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
51
Preglednica 6. Dimenzije in nagibi manjših klifov.
Table 6. Size and slope of smaller cliffs.
nagib / slope
klif
cliff
A
z
 [m
2
] A
y
 [m
2
] nasuti del
deposition area
erodirajo či del
eroding area
Rokava-2 320 310 0.7 1.5
Rokava-3 50 150 - 3.0
Lopar 360 310 0.75 1.8
1.11.00 1.1.01 1.3.01 1.5.01 1.7.01 1.9.01 1.11.01 1.1.02 1.3.02 1.5.02
-20
0
20
40
60
80
100
120
140
dz [mm]
datum
 sr
 max
 min
Slika 6. Napredovanje sedimentacije v žlebi čih (n = 8).
Figure 6. Dynamics of sedimentation in rill area (n = 8).
3.2.4 MERITVE Z ŽEBLJI ČI NA KLIFU
ROKAVA-1
Napredovanje sedimentacije v žlebi čih je
predstavljeno na sliki 6. Iz slike 6 vidimo, da
je klif v obdobju od maja do julija tako reko č
neaktiven. Najve čja sprememba se zgodi z
nastopom pozno poletnih oziroma zgodnje
jesenskih nalivov z veliko erozivno mo čjo
3.2.4 MONITORING WITH EROSION PINS
AT ROKAVA-1 CLIFF
The dynamics of the sedimentation in the
rill area is shown in Figure 6. From Figure 6 it
can be seen that the cliff is practically inactive
in the months of May and June. Changes are
fastest in late summer and early autumn and
are caused by intensive rainfalls (Petkovšek,

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
52
(Petkovšek, 2002). Takrat se na novo pojavijo
žlebi či, v katerih se nato zlasti v novembru in
decembru, pa tudi še v pomladanskih mesecih,
znova odlagajo sedimenti, dokler se žlebi č ne
izravna z okoliško površino. Na podlagi
meritev smo ugotovili, da so žlebi či globoki do
10 cm, široki okrog 20 cm, dolžina pa je
odvisna od velikosti nasutega dela klifa, kjer
se pojavljajo. Tipična dolžina je med 5 in 10
m.
3.3 EROZIVNOST PADAVIN
V obravnavanem obdobju od se je na
obravnavnem obmo čju pojavilo sedemnajst
velikih dogodkov, ki so na vsaj eni lokaciji
presegli mejo R = 100 MJ/ha mm/h. Slika 7 za
te dogodke prikazuje erozivnost padavin po
lokacijah. Preglednica 8 podaja višino padavin
P, najve čjo polurno intenziteto padavin I
30
 in
erozivnost padavin R za deset najve čjih
dogodkov.
Iz rezultatov vidimo, da so se nalivi z
veliko erozivnostjo pojavljali pozno poleti in
jeseni (julij–november). Zunaj tega obdobja, to
je v desetih mesecih od decembra 2000 do
junija 2001 in od decembra 2001 do februarja
2002, sta se pojavila le dva velika dogodka.
Poletni erozivni dogodki so prostorsko
omejenega zna čaja, če upoštevamo, da je
razdalja med dežemeri le nekaj kilometrov.
Tako je bil dogodek 10. 7. 2001 eroziven tako
reko č le v Kocjan či čih, 29. 7. 2001 pa
predvsem v Koštaboni. Tudi pri ostalih
dogodkih je opazno izrazito prostorsko
spreminjanje erozivnosti padavin.
Posebno pozornost zasluži dogodek 31. 8.
2001, ko je najve čja polurna intenziteta
padavin I
30
 presegla 100 mm/h. Podobne in
ve čje vrednosti so sicer zna čilne za poletne in
jesenske nalive na obmo čju Sredozemlja.
Martinez Casanovas et al. (2002) navajajo
dogodek iz osrednje Katalonije (Španija)
junija 2000, ko je najve čja polurna intenziteta
padavin dosegla I
30 
= 170 mm/h, erozivnost
dogodka pa kar R = 11.756 MJ/ha mm/h.
2002). Rills are formed at that period of year.
From then on, the sediment is deposited in rills
(especially in November, December) till they
are filled and levelled with the rest of the cliff
surface in spring. Measurements showed that
the rills are up to 10 cm deep and 20 cm wide
while the length depends on the size of the
deposition area where a rill is formed. Typical
lengths are between 5 and 10 m.
3.3 RAINFALL EROSIVITY
In the monitored area and the period of
observation, seventeen big events occurred
that exceeded the limit of R = 100 MJ/ha
mm/h in at least one location. Figure 7 shows
the value of R factor for these events for all
locations. Table 8 gives precipitation P,
maximum 30 minute intensity I
30
 and R factor
for ten biggest events.
The results indicate that the storms with
high erosivity occur in late summer and in
autumn (July–November). Outside of that
period, i.e. in ten months between December
2000 and June 2001 and December 2001 and
February 2002, only two big events occurred.
During the summer, erosive events are of
spatially limited character, considering that the
distance between the raingauges is only a
couple of kilometres. The event of July 10,
2001 was limited to Kocjan či či, while the
event of July 29, 2001 was more or less
limited to Koštabona. A similar pattern could
be observed at other events.
A special attention must be paid to the
event of August 31, 2001, when the maximum
30 minute intensity I
30
 exceeded 100 mm/h.
Similar values are characteristic for summer
storms in the Mediterranean. Martinez
Casanovas et al. (2002) inform about an event
from central Catalonia (Spain), where in June
2000 the following values were recorded: I
30
 =
170 mm/h and R = 11756 MJ/ha mm/h.

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
53
0 500 1000 1500 2000 2500 3000
10-okt-00
17-okt-00
3-nov-00
4-nov-00
6-nov-00
8-nov-00
15-nov-00
17-nov-00
16-apr-01
31-maj-01
10-jul-01
20-jul-01
29-jul-01
31-avg-01
14-sept-01
4-okt-01
12-nov-01
datum / date
R [MJ/ha mm/h]
Koštabona
Dragonja
Rokava
Marezige
Kocjan či či
Slika 7. Pregled dogodkov z veliko erozivnostjo (R > 100 MJ/ha mm/h).
Lokacije so razvrš čene v obratni smeri vodnega toka.
Figure 7. Overview of the events with high erosivity (R > 100 MJ/ha mm/h).
Locations are ordered in the upstream direction.

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
54
Preglednica 8. Pregled desetih najve čjih erozivnih dogodkov.
Table 8. Overview of ten biggest erosive events.
Št. / No
datum / date
d.m.yy
Lokacija
Location
P
 mm
I
30
mm/h
R
 MJ/ha mm/h
1 31.8.2001 Rokava 86.5 108.8 2647
2 31.8.2001 Dragonja 74.2 85.5 1759
3 14.9.2001 Marezige 105.2 54.4 1368
4 31.8.2001 Marezige 62.8 79.7 1366
5 14.9.2001 Rokava 115.6 44.8 1267
6 4.11.2000 Rokava 61.2 68.6 1081
7 14.9.2001 Kocjan či či 105.0 42.1 1022
8 31.8.2001 Kocjan či či 68.2 50.7 853
9 4.11.2000 Dragonja 54.4 54.7 736
10 10.10.2000 Koštabona 84.9 32.6 558
4. ZAKLJU ČEK
Na podlagi opravljenih meritev od jeseni
2000 do pomladi 2002 lahko povpre čni letni
odtok lebde čih plavin s povodja Rokave
ocenimo na okrog 1500 t (750 kg/ha),
povpre čni letni odtok rinjenih plavin pa na
vsaj 10 % te vrednosti.
Ocenjujemo, da je premeš čanje ve čjih zrn
rinjenih plavin omejeno v povpre čju na en
dogodek letno. Rinjene plavine izvirajo
predvsem iz klifov in morda bo čne erozije že
odloženega materiala.
Rezultati kažejo, da je vir ve čjega dela
lebde čih plavin površinska erozija, manjšega
pa klifi (nekaj 100 m
3
 oz 100 ton letno). Za
natan čnejšo dolo čitev deležev so potrebne
dolgotrajnejše meritve.
Erozivni dogodki so zelo neenakomerno
razporejeni, tako časovno kot prostorsko.
Najve čji dogodki se pojavljajo pozno poleti in
jeseni. V tem času je tudi erozijska aktivnost
klifov najve čja.
Pri meritvah klifov se je za uspešno
izkazala kombinacija fotogrametri čne meritve
in meritve z erozijskimi žeblji či.
4. CONCLUSION
The following conclusions apply to the
measurements in Rokava catchment from
autumn 2000 to spring 2002. The average
annual suspended sediment yield is around
1500 t (750 kg/ha).
The average annual bedload transport is
about one tenth of that value and on average
occurs during one event per year. The source
of bedload is rocks that originate from cliffs
and possibly bed and bank material.
The results show that the source of
suspended sediment is mostly surface erosion,
while cliffs contribute a couple 100 m
3
 or
approximately 100 tonnes per year. Further
measurements in a longer period are required
to obtain more precise results.
Erosive events are highly variable both
temporally and spatially. The biggest events
usually occur in late summer and in autumn.
During this time, the activity of the cliffs is
also at its peak.
For monitoring of cliffs, the combination of
photogrammetry and erosion pins proved to be
successful.

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
55
VIRI – REFERENCES
Gavrilovi ć, S. (1970). Savremeni na čini prora čunavanja buji čnih nanosa i izrada karata erozije.
(Contemporary methods for calculation of sediment yield and soil erosion mapping).
Proceedings of Erosion, torrents and sediments. Institut Jaroslav Černi, Belgrade, 85–100 (in
Serbo-Croatian).
GI (2001). 3D Model klifa ob Rokavi (3D model of the Rokava cliff). Report on photogrammetric
survey, Geodetic institute of Slovenia, Ljubljana, 15 p.
GI (2002a). 3D Model klifa ob Rokavi (3D model of the Rokava cliff). Report on photogrammetric
survey, Geodetic institute of Slovenia, Ljubljana, 15 p.
GI (2002b). 3D Model klifa ob Rokavi, tretja izmera (3D model of the Rokava cliff, third
measurement). Report on photogrammetric survey, Geodetic Institute of Slovenia, Ljubljana, 15
p.
Globevnik, L., Sovinc, A. (1998). The impact of catchment land use change on river flows: the
Dragonja river, Slovenia. Hydrology in a changing environment I. John Wiley & Sons, 525–533.
Globevnik, L. (2001). Celosten pristop k urejanju voda v povodjih (Integrative approach to water
management) Unpublished Doctoral thesis, University of Ljubljana, 167 pp.
Hahn, C.T., Barfield, B.J., Hayes, J.C. (1994). Design hydrology and sedimentology for small
catchments. Academic Press Inc., San Diego, ZDA, 588 pp.
Hobby, D., Minneboo, S. (2001). Hydrology of upper Dragonja catchment, SW Slovenia, a
preliminary analysis. Unpublished document of VU Amsterdam, 101 p.
Jurak, V., Petraš, J., Gajski, D. (2002). Istraživanje ekscesivne erozije na ogoljenim flišnim
padinama u Istri primjenom trestri čke fotogrametrije. (Survey of excessive soil erosion from bare
flysch slopes in Istria by terrestrial photogrammetry). Hrvatske vode 10/38, 49–58 (in Croatian).
Leica Geosystems (2001). DISTO pro User Manual, 242 p.
Martínez Casanovas, J.A., Ramos, M.C., Ribes-Dasi, M. (2002). Soil erosion caused by extreme
rainfall events: mapping and quantification in agricultural plots from very detailed digital
elevation models. Geoderma 105, 125–140.
Meyer, L.D., Wischmeier, W.H. (1969). Mathematical simulation of the process of soil erosion by
water. Trans. ASAE 12(6), 754–758.
Petkovšek, G. (2000). Procesno utemeljeno modeliranje erozije tal = Process based soil erosion
modelling. Acta hydrotechnica 18/28, 41–60.
Petkovšek, G. (2002). Modeliranje in kvantifikacija erozije tal z aplikacijo na povodju Dragonje
(Modelling and quantification of soil erosion with application to Dragonja catchment, in
Slovenian). Doctoral thesis, University of Ljubljana, 205 p.
Pintar, J., Mikoš, M. (1983). Izdelava smernic in normativov z globalno usmeritvijo urejanja po
ekosistemih, pojavnostih in ekološki primernosti ter na činov gospodarjenja s povirji voda.
Poro čilo VGI C-432, Ljubljana, 133 p.
PUH (1971). Erozija tal in hudourniki, Dragonja v slovenski Istri (Soil erosion and torrents,
Dragonja in Slovenian Istria). Report, Ljubljana (in Slovenian).
Reid, L.M., Dunne, T. (1996). Rapid Evaluation of Sediment Budgets. Catena Verlag, 164 p.
Renard, K.G., Foster, G.R., Weesies, G.A., McCool, D.K., Yoder, D.C. (1997). Predicting soil
erosion by water: a guide to conservation planning with the Revised universal soil loss equation
(RUSLE). USDA, Agricultural Handbook No 703, 404 p.

Petkovšek, G., Mikoš, M.: Erozijski procesi v eksperimentalnem povodju Dragonje, JZ Slovenija – Measurements of
Erosion Processes in the Experimental Catchment of the Dragonja River, SW Slovenia
© Acta hydrotechnica 21/34 (2003), 37–56, Ljubljana
56
Naslovi avtorjev – Authors' Addresses
asist. dr. Gregor Petkovšek
Univerza v Ljubljani – University of Ljubljana
Fakulteta za gradbeništvo in geodezijo – Faculty of Civil and Geodetic Engineering
Jamova 2, SI – 1000 Ljubljana
E-mail: gpetkovs@fgg.uni-lj.si
present address:
CGS, Ra čunalniško podprto projektiranje, GIS in ekologija, d.o.o. –
CGS, Computer Aided Design, GIS and Ecology, L.t.d.
Brn či čeva 13, SI-1000 Ljubljana
E-mail: Gregor.Petkovsek@cgsplus.si
izr. prof. dr. Matjaž Mikoš
Univerza v Ljubljani – University of Ljubljana
Fakulteta za gradbeništvo in geodezijo – Faculty of Civil and Geodetic Engineering
Jamova 2, SI – 1000 Ljubljana
E-mail: mmikos@fgg.uni-lj.si