© Acta hydrotechnica 22/37 (2004) Ljubljana
ISSN 1581-0267
95
UDK/UDC: 519.61/.64:504.4:556.38 Prejeto/Received: 18. 6. 2005
Kratki znanstveni prispevek – Preliminary scientific paper Sprejeto/Accepted: 25. 11. 2005
UPORABA MODELA PEARL ZA SIMULACIJO EMISIJE PESTICIDOV V
PODTALNICO LJUBLJANSKEGA POLJA
USAGE OF MODEL PEARL FOR SIMULATION OF PESTICIDE EMISSION
INTO THE LJUBLJANSKO POLJE GROUNDWATER
Sašo KLEMENČI Č
Podtalnica Ljubljanskega polja je eno najpomembnejših vodnih teles v Sloveniji in predstavlja vir
pitne vode za okoli 300.000 ljudi. Tla, ki prekrivajo vodonosnik, so ve činoma plitva in dobro
prepustna, zaradi tega podtalnico ogrožajo pesticidi, ki pronicajo iz kmetijskih površin na
Ljubljanskem polju. Velike koli čine v preteklosti uporabljenega atrazina na Ljubljanskem polju so
privedle do tega, da sta atrazin in njegov razgradni produkt desetil atrazin presegla mejne
vrednosti in sta še vedno prisotna v podtalnici. Zaradi manjše porabe po letu 1995 in prepovedi
uporabe atrazina leta 2003 se njuna vrednost po časi zmanjšuje. Da bi se v prihodnosti takšna
stanja prepre čilo, smo skušali ugotoviti uporabnost modela PEARL za ocenjevanje bilance atrazina
in mobilnosti pesticidov, ki se danes uporabljajo.
Klju čne besede: podtalnica, atrazin, metolaklor, PEARL, Ljubljansko polje
Groundwater of the plain of Ljubljansko polje is one of the most important water bodies in Slovenia
and represents the source of drinking water for around 300,000 people. Soil covering the aquifer is
mostly shallow and permeable. Because of that the percolation of pesticides from the Ljubljansko
polje fields represents a considerable threat to the groundwater. In the past atrazine was used in
large amounts at the Ljubljansko polje, which was the reason that atrazine and its metabolite
desethyl atrazine have exceeded the limit value and are still present in the groundwater. After 1995,
the use of atrazine started to decrease, and in 2003 the use of atrazine was prohibited. Because of
that the quantity of atrazine and desethyl atrazine in groundwater is slowly reducing. In order to
prevent such scenarios in the future, we tried to establish the suitability of the PEARL model for the
Ljubljansko polje. With the model one can evaluate the pesticide mobility, before they are
registered and deployed to fields.
Key words: groundwater, atrazine, metolaclor, PEARL, Ljubljansko polje
1. UVOD
Uporaba razli čnih kemi čnih sredstev za
zaš čito rastlin, ki se jih uporablja v kmetijstvu,
onesnažuje okolje. Ker gre za strupene snovi,
je njihova prisotnost v okolju in še posebej v
pitni vodi nevarna. Ko pesticidi z dolgo
razpolovno dobo pridejo do podtalnice, se
dolgo časa zadržijo v njej, pri tem po časi
razpadajo in prehajajo v druge, tudi strupene
snovi. Lahko samo po čakamo, da se odstranijo
po naravni poti, kar pa traja dolgo časa, ogroža
zdravje ljudi ter pove čuje stroške oskrbe s
pitno vodo. Ker je treba takšne primere
prepre čiti, smo preverjali uporabnost modela
1. INTRODUCTION
The usage of different chemical agents for
plant protection used in farming has been the
cause of environmental pollution. Because of
toxicity of chemical agents their presence in
the environment, especially in drinking water,
is hazardous. Once pesticides with long half
lives enter the groundwater, they stay there for
a long time, slowly breaking down and
transforming themselves into other also toxic
substances. We have no other choice but to
wait for the pesticides to break down by
themselves. However, that takes a long time,
represents health risks to people and increases
the costs of water supply. Such situations need
to be prevented. Because of that we examined
the PEARL model (Pesticide Emission

Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
96
PEARL (Pesticide Emission Assessment at
Regional and Local Scales) za ugotavljanje
mobilnosti pesticidov pri pogojih, kakršni
obstajajo na Ljubljanskem polju. Uporabnost
je bilo treba preveriti, ker je bil model razvit
na Nizozemskem in je prilagojen njihovim
razmeram. Z uporabo takih modelov bi se
lahko pred registracijo pesticidov simuliralo
njihovo obnašanje v okolju in bi se tako
prepre čilo kontaminiranje podtalnice.
Vodonosnik Ljubljanskega polja je eden
najve čjih in tudi najpomembnejših vodnih
teles v Sloveniji. Tla, ki ga prekrivajo, so
plitva in dobro prepustna, zato je tudi eden
izmed najbolj ogroženih vodonosnikov v
Sloveniji. Njegove dinami čne zaloge so
ocenjene na 3 do 4 m
3
/s (Andjelov et al.,
2005). Njegove stati čne zaloge ob povpre čnem
vodostaju podtalnice pa po oceni znašajo
1,801 x 10
12 
m
3 
. Kljub tako velikim koli činam
vode so bile v nekaterih vodnjakih presežene
mejne vrednosti za atrazin – v vodarni Hrastje
pa so še vedno. To je vodilo k temu, da se je
uporaba atrazina za čela po letu 1995
zmanjševati, leta 2003 pa je bila sprejeta
prepoved uporabe atrazina.
2. MATERIAL IN METODE
Pesticidi, ki se uporabljajo v kmetijstvu, se
lahko izcejajo v podtalnico ali izhlapevajo v
ozra čje. Za ocenjevanje nevarnosti emisij
pesticidov iz rastlin in tal ter njihovo
obnašanje v okolju se vse bolj uporabljajo
ra čunski modeli. Izbrali smo model PEARL in
preverili njegovo uporabnost za vodonosnik
Ljubljanskega polja (Leistra et al., 2001).
PEARL (slika 1) je enodimenzionalni
numeri čni model, ki simulira obnašanje
pesticidov v sistemu zemljina–rastlina. Razvit
je bil na Nizozemskem. Tok vode po zemljini
je opisan z Richardovo ena čbo, upoštevajo č
vrsto možnih spodnjih robnih pogojev.
Evaporacija zemljine in transpiracija rastlin sta
izra čunana z množenjem referen čne
evapotranspiracije s faktorji rastlin in zemljine.
Tok toplote po zemljini je opisan s
Fourierjevim zakonom. Toplotne lastnosti so
funkcija poroznosti in vodne vsebnosti in so
zato funkcija časa in globine zemljine.
Assessment at Regional and Local Scales) for
simulating pesticide mobility in conditions
present in the plain of Ljubljansko polje. Since
the model has been developed in the
Netherlands and adapted to their conditions,
we had to check the applicability of the model
in Slovenia. Namely, prior to the actual
registration and application of pesticides to
cropland, we could, with the help of these
types of models, simulate the pesticide
behaviour in the environment. In this way the
contamination of groundwater could be
prevented.
The aquifer of the Ljubljansko polje is one
of the biggest and most important water bodies
in Slovenia. The soil covering the Ljubljansko
polje aquifer is shallow and permeable, which
is why this is one of the most vulnerable
aquifers in Slovenia. Its dynamic reserves
were estimated at 3–4 m
3
/s (Andjelov et al.,
2005), whereas its static reserves at the
average water level of groundwater were
estimated at 1.801 x 10
12 
m
3
. Despite such
enormous quantity of water in some wells, the
concentrations of atrazine were over the
boundary value. In the Hrastje water plant, the
concentrations are still over the boundary
value. Because of high concentrations of
atrazine in groundwater its usage has begun to
decrease after 1995; since 2003 the usage of
atrazine has been prohibited.
2. MATERIAL AND METHODS
Pesticides used for farming can leach into
underground waters or evaporate into the
atmosphere. For evaluating the danger of
pesticide emission from plants and soil,
numerical models are used. We chose the
PEARL model and examined its applicability
for the Ljubljansko polje aquifer (Leistra et al.,
2001).
PEARL (Fig. 1) is a one-dimensional
numerical model, which simulates pesticide
behaviour in the soil–plant system. It was
developed in the Netherlands. The water flow
in soil is described by Richard’s equation
including a range of possible lower boundary
conditions. Evaporation from the soil and
transpiration from plants are calculated so that
the reference evapotranspiration is multiplied
by soil and plant factors. Soil heat flow is
described by Fourier’s law. The thermal
properties are a function of porosity and water
content and are therefore a function of time
and soil depth.

Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
97
Model PEARL temelji na:
- konvekcijsko/disperzijski ena čbi,
vklju čujo č difuzijo v plinasti fazi s
temperaturno odvisnim Henryjevim
koeficientom,
- modelu Freundlichove absorpcije,
- hitrosti transformacije, ki je odvisna od
vsebnosti vode, temperature in globine
zemljine,
- pasivnem rastlinskem dvigovanje vode.
Model ra čuna tudi tvorbo in obnašanje
produktov transformacije in opisuje tudi
stranski izpust pesticidov v drenaže. Model ne
simulira preferen čnega toka. Izhlapevanje iz
površja zemljine je izračunano tako, da se
privzame laminarno zra čno plast na površini
zemljine (Leistra et al., 2001).
PEARL sam ne ra čuna toka vode in
temperatur zemljine, ampak uporabi izra čune
iz modela SWAP (Soil Water Atmosphere
Plant). Programski paket za simulacijo je torej
sestavljen iz dveh modelov: SWAP in PEARL
(slika 2) (Leistra et al., 2001).
Simulirali smo nanašanje atrazina na
Ljubljansko polje, in sicer v realnih koli činah,
ki so jih ocenili na Inštitutu za hmeljarstvo in
pivovarstvo Žalec (Simon či č et al., 2004).
Privzeli smo, da so atrazin nanašali vsako leto
na isto polje, poraš čeno s koruzo. Takega
stanja v realnosti zaradi kolobarjenja ni.
Koli čine in datumi nanosov, uporabljeni v
modelu, so prikazani v preglednici 1.
The PEARL model is based on:
- the convection/dispersion equation
including diffusion in the gas phase
with the temperature dependent on the
Henry coefficient,
- the Freundlich sorption model,
- the transformation rate that depends on
water content, temperature and depth in
soil,
- the passive plant uptake rate.
PEARL also calculates the formation and
behaviour of transformation products and
describes the lateral discharge of pesticides
into drains. The volatilisation from the soil
surface is calculated with the assumption that
there is a laminar air layer on the soil surface
(Leistra et al., 2001).
The PEARL model does not calculate water
flow and soil temperatures but uses calculated
values from the SWAP model (Soil Water
Atmosphere Plant). The program package for
the simulation consists of two models: SWAP
and PEARL (Fig. 2) (Leistra et al., 2001).
We simulated the application of atrazine to
the Ljubljansko polje and took real quantities
that were estimated by the Slovenian Institute
of Hop Research and Brewing (Simon či č et
al., 2004). We assumed that the atrazine was
applied to the same cornfield every year. In
reality such a condition does not exist because
of the rotation of crops. Quantities and dates of
application used in the model are represented
in Table 1.
Preglednica 1. Nanašanje atrazina na polje v simulaciji.
Table 1. Application of atrazine to the field in simulation.
Datum / Date Nanos / Application
(kg/ha)
Datum / Date Nanos / Application
(kg/ha)
13.05.1989 1.500 06.05.1997 0.833
11.05.1990 1.500 13.05.1998 0.500
14.05.1991 1.500 08.05.1999 0.300
11.05.1992 1.500 07.05.2000 0.233
07.05.1993 1.500 15.05.2001 0.167
10.05.1994 1.500 08.05.2002 0.067
10.05.1995 1.500 2003 0.000
07.05.1996 1.330 2004 0.000

Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
98
Slika 1. Pregled podatkovne baze modela PEARL (Tiktak et al., 2000).
Figure 1. Overview of the PEARL database (Tiktak et al., 2000).

Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
99
Slika 2: Pregled procesov, vklju čenih v model PEARL (Tiktak et al., 2000).
Figure 2. Overview of processes included in the PEARL model (Tiktak et al., 2000).

Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
100
Uporabljeni so bili realni meteorološki
podatki, kot so: son čno sevanje, maksimalna
dnevna temperatura, minimalna dnevna
temperatura, povpre čni parni tlak, hitrost vetra
in padavine od 1. 1. 1989 do 30. 9. 2004
(ARSO). V simulaciji smo uporabili troje
razli čnih tal: lahka tla, lahka in srednje težka
tla in srednje težka tla. Lahka tla ležijo ob reki
Savi in predstavljajo podro čje, kjer je
ogroženost podtalnice zaradi uporabe
fitofarmacevtskih sredstev najve čja. Ve čji del
teh tal prekrivajo gozdovi, grmi čevje in
travniki. Najve čji del prispevnega podro čja, iz
katerega se napajajo vodnjaki v vodarnah, je
prekritega z lahkimi in srednje težkimi tlemi.
Srednje težka tla so prisotna le na zelo
majhnem delu prispevnega obmo čja. Podatki o
lastnostih tal so podani v preglednicah 2–9.
Simulirali smo tudi nanašanje atrazina na
železniške nasipe (podatki o nanašanju so isti
kot pri poljedelstvu).
Za primerjavo smo simulirali tudi
distribucijo metolaklora po zemljini.
Metolaklor smo izbrali, ker se je uporabljal
približno v istih koli činah kot atrazin, vendar
pa mejnih vrednosti v vodnjakih vodarn ni
nikoli presegel (Andjelov et al., 2005).
Lastnosti pesticidov atrazina in metolaklora,
uporabljene v izra čunu, so navedene v
preglednicah 10 in 11.
The applied meteorological data were real;
they were obtained from the Environmental
Agency of the Republic of Slovenia (ARSO).
We entered the following data: daily global
sun radiation, maximum daily temperature,
minimum daily temperature, average vapour
pressure, average wind speed and daily
precipitation, dating from 01/01/1989 to
30/09/2004. In the simulation we used three
types of soil: light soil, light and moderate soil,
and moderate soil. Light soil is to be found
near the river Sava and represents the area
where the groundwater is most vulnerable to
the use of pesticides. The major part of light
soil is covered with woods, bushes and
meadows. The largest part of the drainage
basin is covered with light and moderate soil.
Moderate soil is present only in a small part of
the drainage basin. Data about soil properties
are given in Tables 2 to 9. We also simulated
the application of atrazine to railway dikes
(application data are the same as for farming).
For comparison we also simulated soil
distribution of metolaclor. Metolaclor was
chosen because it was used in almost the same
quantities as the atrazine, however, the
concentrations of metolaclor in the wells of
water plants were never exceeded (Andjelov et
al., 2005). Properties used in the model for
atrazine and metolaclor are given in Tables 10
and 11.
Preglednica 10. Razpolovni čas atrazina v dneh v odvisnosti od vrste tal in globine.
Table 10. Half life of atrazine in days, as a function of soil type and depth.
Horizont
Horizon
123456
Lahka tla
Light soil
40 146 300 1000 1000 –
Lahka in srednje težka tla
Light and moderate soil
40 146 300 1000 1000 –
Srednje težka tla
moderate soil
40 146 146 300 1000 1000
Železniški nasipi
Railway dikes
146 1000 1000 – – –

Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
101
Suha pros. gostota zem.
Dry bulk density
[kg/m
3
]
1260
1260
1860
2000
2000
Glina
Clay
[%]
6
15
6
2.5
2.5
Pesek
Sand
[%]
46
33
90
95
95
Organska snov
Organic matter
[%]
5
3.25
1
0
0
pH
8
8
8
7
7
Teksturni razred
Texture class
PI – peš čena ilovica
SaL – Sandy Loam
MI – meljasta ilovica
SiL – Silty Loam
prod in pesek
Gravel and Sand
prod in pesek
Gravel and Sand
prod in pesek
Gravel and Sand
Debelina
plasti
Layer
thickness
[cm]
15
14
50
200
721
Preglednica 2. Parametri, uporabljeni v modelu PEARL, za lahka tla.
Table 2. Parameters used in PEARL for light soil.
1. horizont/horizon 1
2. horizont/horizon 2
3. horizont/horizon 3
4. horizont/horizon 4
5. horizont/horizon 5
λ (–)
0.18
0.057
0
0
0
K
sat
 (m/d)
10
3.0912
8
10
10
n (–)
1.56
1.686
2.17
2.17
2.17
α
w
 (1/cm)
0.016
0.0025
0.022
0.022
0.022
α
d
 (1/cm)
0.016
0.0025
0.022
0.022
0.022
θ
s
 (m
3
/m
3
)
0.01
0.00
0.01
0.01
0.01
θ
s
 (m
3
/m
3
)
0.524
0.523
0.379
0.323
0.323
Preglednica 3. Uporabljeni Van Genuchtenovi parametri za lahka tla
(Tiktak et al. , 2000: K
sat 
iz SWCHPC, 2004).
Table 3. Van Genuchten parameters used for light soil
(Tiktak et al., 2000: K
sat 
from SWCHPC, 2004).
 
1. horizont (TOP Soils – B4)
  2. horizont (SUB Soils – O14)
3. horizont (SUB Soils – O1)
4. horizont (SUB Soils – O1)
5. horizont (SUB Soils – O1)

Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
102
Suha pros. gostota
Dry bulk density
[kg/m
3
]
1100
1100
1860
2000
2000
Glina
Clay
[%]
23
25
6
2.5
2.5
Pesek
Sand
[%]
16
18
90
95
95
Organska snov
Organic matter
[%]
7
5
1
0
0
pH
6
6
7
7
7
Teksturni razred
Texture class
MI – meljasta ilovica
SiL – Silty Loam
MI – meljasta ilovica
SiL – Silty Loam
prod in pesek
Gravel and Sand
prod in pesek
Gravel and Sand
prod in pesek
Gravel and Sand
Debelina plasti
Layer thickness
[cm]
20
16
50
200
724
Preglednica 4. Parametri, uporabljeni v modelu PEARL, za lahka in srednje težka tla (iz Simon či č et al., 2004).
Table 4. Parameters used in PEARL for light and moderate soil (from Simonči č et al., 2004).
1. horizont/horizon 1
2. horizont/horizon 2
3. horizont/horizon 3
4. horizont/horizon 4
5. horizont/horizon 5
λ (–)
–2.161
0.057
0
0
0
K
sat
 (m/d)
6.0648
4.8504
8
10
10
n (–)
1.325
1.686
2.167
2.167
2.167
α
w
 (1/cm)
0.0065
0.0025
0.0224
0.0224
0.0224
α
d
 (1/cm)
0.0065
0.0025
0.0224
0.0224
0.0224
θ
s
 (m
3
/m
3
)
0
0.00
0.01
0.01
0.01
θ
s
 (m
3
/m
3
)
0.584
0.586
0.379
0.323
0.323
Preglednica 5. Uporabljeni Van Genuchtenovi parametri za lahka in srednje težka tla
(Tiktak et al., 2000: K
sat 
iz SWCHPC, 2004).
Table 5. Van Genuchten parameters for light and moderate soil (Tiktak et al., 2000: K
sat 
from SWCHPC, 2004).
1. horizont (TOP Soils – B09)
2. horizont (SUB Soils – O14)
3. horizont (SUB Soils – O1)
4. horizont (SUB Soils – O1)
5. horizont (SUB Soils – O1)

  Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
103
Suha pros. gostota zem.
Dry bulk density
[kg/m
3
]
1100
1100
1210
1310
2000
2000
Glina
Clay
[%]
23
25
22
40
2.5
2.5
Pesek
Sand
[%]
16
18
22
19
95
95
Organska snov
Organic matter
[%]
7
5
3
0
0
0
pH
6
6
6
6
6.5
7
Teksturni razred
Texture class
MI – meljasta ilovica
SiL – Silty Loam
MI – meljasta ilovica
SiL – Silty Loam
MI – meljasta ilovica
SiL – Silty Loam
MG – meljasta glina
SiC – Silty Clay
prod in pesek
Gravel and Sand
prod in pesek
Gravel and Sand
Debelina plasti
Layer thickness
[cm]
20
16
29
15
200
720
Preglednica 6. Parametri, uporabljeni v modelu PEARL, za srednje težka tla.
Table 6. Parameters used in the PEARL model for moderate soil.
1. horizont/horizon 1
2. horizont/horizon 2
3. horizont/horizon 3
4. horizont/horizon 4
5. horizont/horizon 5
6. horizont/horizon 6
λ (–)
–2.161
0.057
0.057
–4.171
0
0
K
sat
 (m/d)
6.0648
4.8504
1.6848
0.0312
10
10
n (–)
1.325
1.686
1.686
1.159
2.167
2.167
α
w
 (1/cm)
0.0065
0.0025
0.0025
0.0095
0.0224
0.0224
α
d
 (1/cm)
0.0065
0.0025
0.0025
0.0095
0.0224
0.0224
θ
s
 (m
3
/m
3
)
0
0.00
0
0
0.01
0.01
θ
s
 (m
3
/m
3
)
0,584
0.586
0.542
0.508
0.323
0.323
Preglednica 7. Uporabljeni Van Genuchtenovi parametri za srednje težka tla
(Tiktak et al., 2000: K
sat 
iz SWCHPC, 2004).
Table 7. Van Genuchten parameters used for moderate soil
(Tiktak et al., 2000: K
sat 
from SWCHPC, 2004).
 
 1. horizont (TOP Soils – B09)
 2. horizont (SUB Soils – O14)
 3. horizont (SUB Soils – O14)
 4. horizont (SUB Soils – O12)
 5. horizont (SUB Soils – O1)
 6. horizont (SUB Soils – O1)

  Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
104
Suha pros. gostota zem.
Dry bulk density
[kg/m
3
]
1860
2000
2000
2000
Glina
Clay
[%]
6
2.5
2.5
2.5
Pesek
Sand
[%]
90
95
95
95
Organska snov
Organic matter
[%]
2
0
0
0
pH
8
7
7
7
Teksturni razred
Texture class
prod in pesek / Gravel and Sand
prod in pesek / Gravel and Sand
prod in pesek / Gravel and Sand
prod in pesek / Gravel and Sand
Debelina
plasti
Layer
thickness
[cm]
10
90
200
700
Preglednica 8. Parametri, uporabljeni v modelu PEARL, za železniški nasip.
Table 8. Parameters used in model PEARL for railway dike.
1. horizont/horizon 1
2. horizont/horizon 2
3. horizont/horizon 3
4. horizont/horizon 4
λ (–)
0
0
0
0
K
sat
(m/d)
8
10
10
10
n (–)
2.167
2.167
2.167
2.167
α
w
(1/cm)
0.0224
0.0224
0.0224
0.0224
α
d
(1/cm)
0.0224
0.0224
0.0224
0.0224
θ
s
(m
3
/m
3
)
0.01
0.01
0.01
0.01
θ
s
(m
3
/m
3
)
0.379
0.353
0.353
0.353
Preglednica 9. Uporabljeni Van Genuchtenovi parametri za železniški nasip (Tiktak et al., 2000: K
sat 
iz SWCHPC, 2004).
Table 9. Van Genuchten parameters for railway dike (Tiktak et al., 2000: K
sat 
from SWCHPC, 2004).
 
 1. horizont (SUB Soils – O1)
 2. horizont (SUB Soils – O1)
 3. horizont (SUB Soils – O1)
 4. horizont (SUB Soils – O1)

Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
105
Preglednica 11. Absorpcija atrazina v odvisnosti od zemljine – K
oc
 iz (ARS PPD, 2005).
Table 11. Sorption of atrazine in dependence of soil – K
oc 
from (ARS PPD, 2005).
K
oc
K
om
peš čena ilovica (PI) / Sandy Loam (SaL) 57 33
meljasta ilovica (MI) / Silty Loam (SiL) 120 70
glina (G) / Clay (C) 87 50
pesek (P) / Sand (Sa) 90 52
Simulirali smo precedek atrazina in
metolaklora na globini 10 m (najve čja globina,
ki jo podpira model PEARL), kar pa ni dovolj
globoko, saj podtalnica Ljubljanskega polja
ponekod leži tudi 30 m pod površjem. Vendar
so izra čuni pokazali, da se atrazin v globljih
plasteh, kjer ni veliko organskih snovi in
mikroorganizmov, skoraj ne razkraja, zato
izra čunani rezultati dajejo dober približek
dejanskim procesom.
3. REZULTATI IN RAZPRAVA
Dobljeni rezultati (slika 3) so pokazali, da
so tla na Ljubljanskem polju zaradi svoje
plitvosti zelo slaba zaš čita podtalnice, saj
zadržijo večjo koncentracijo atrazina samo v
zgornji plasti tal (lahka tla do 29 cm, lahka in
srednje težka tla do 36 cm in srednje težka tla
do 65 cm). Ko pa atrazin doseže globino 1 m,
se veže na tla samo v zelo majhnih koli činah,
ostalo pa pronica proti podtalnici.
Rezultati so pokazali tudi (slika 4), kako z
globino pada koncentracija atrazina v
precedku. Najve čji padec koncentracije je v
plasteh nekaj centimetrov pod površjem.
Naslednji padec se zgodi, ko atrazin preide iz
zemljine v prod in pesek. Potem pa
koncentracije v precedku z globino zelo po časi
upadajo. To pomeni, da do podtalnice, ki je
ponekod v globini 30 m, pridejo koncentracije,
ki niso veliko manjše od koncentracij, ki so
bile izra čunane za precedek na globini 10
metrov.
Izra čunali smo tudi precedke na globini
10 m (sliki 5 in 6) za atrazin (za vse podlage)
in metolaklor (za lahka tla).
We simulated the atrazine and metolaclor
percolate at a depth of 10 m (this is the
maximum depth supported by the model),
which is not deep enough, because the
groundwater of the Ljubljansko polje in some
places lies as much as 30 m deep. However,
calculations have shown that the atrazine
decomposes very slowly in deeper layers of
soil where there is almost no organic matter or
microorganisms. Thus the calculated results
give a good approximation for the actual
processes.
3. RESULTS AND DISCUSSION
The results given on Figure 3 have shown
that the shallowness of the soil of the
Ljubljansko polje represents poor protection to
its groundwater. A considerable amount of
atrazine is retained in the top layer of soil:
light soil up to 29 cm, light and moderate soil
up to 36 cm, and moderate soil up to 65 cm.
When the atrazine reaches the depth of 1 m,
only small amounts of atrazine bind to the soil
and the rest leaches into the groundwater.
The results given on Figure 4 show how the
concentration of the atrazine in the percolate
decreases with depth. The most significant fall
of the concentration happens in the layers just
a few centimetres under the surface. The next
fall happens when the atrazine passes from the
soil into sand and gravel. After that the
concentration in the percolate reduces slowly.
It can be assumed that the concentrations in
the percolate at 30 m depth are not much lower
than the concentration at a depth of 10 m.
We also calculated the percolates at 10 m
depth (Figures 5 and 6) for atrazine (for all
soils) and metolaclor (for light soil).

  Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
106
0
1
2
3
4
5
6
7
8
9
10
0.1 1 10 100 1000 10000
Koncentracija / Concentration [ µg/l]
Globina / Depth [m]
13.05.1995
23.06.1995
13.01.1996
06.05.1996
Slika 3. Koncentracija atrazina v lahki in srednje težki zemljini
na razli čnih globinah ob razli čnih časih.
Figure 3. Concentration of atrazine in light and moderate soil
at different depths and at different dates.
V preglednici 12 so prikazane maksimalne
koncentracije v precedku na globini 10 m.
Prikazana je tudi frakcija atrazina, ki je
dosegla podtalnico.
Primerjali smo izra čunane vrednosti za
razli čne tipe tal z izmerjenimi vrednostmi na
merilnih mestih. Pokazalo se je, da obstaja
neka korelacija. Težava je v tem, da so
izmerjene vrednosti podane od 1992 naprej, na
nekaterih mestih celo od leta 1995 naprej. Prva
merjenja so bila izvedena s predolgimi
presledki, zato je podatke težje primerjati. Pri
simulaciji pa je najve čja napaka na njenem
za četku zaradi privzetega za četnega pogoja, da
je koncentracija atrazina v tleh in podtalnici
enaka ni č. Podatki se razlikujejo tudi po tem,
In Table 12 maximum concentrations in the
percolate at a depth of 10 m are shown.
Fractions of atrazine that leached into the
groundwater are also shown.
We compared the calculated results for
different types of soil with values measured in
the wells. The comparison showed some
correlation between the measured and
calculated data. The problem was that
measured data were available only from 1992
onwards, and at some sampling sites even only
from 1995 onwards. These first measurements
were taken in long intervals, consequently it
was hard to compare the data. In a simulation,
however, the greatest mistake occurs at the
beginning because of the assumed initial
condition that the concentration of atrazine in
soil and in the groundwater is zero. The

  Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
107
da za čnejo ra čunane vrednosti po zmanjšanju
porabe atrazina mo čno padati, medtem ko
merjene vrednosti le po časi padajo. Razlog
temu je, da se pri ra čunanju upošteva samo
plast na globini 10 m pod ra čunano površino,
na merjene vrednosti pa vpliva širše prispevno
obmo čje, s katerega se atrazin izceja v
podtalnico. Kot primer bi lahko podal vodarno
Hrastje, ki ima najbolj raztegnjeno prispevno
obmo čje; to se pozna tudi pri izmerjenih
vrednostih, saj so koncentracije atrazina v tej
vodarni še vedno nad dovoljeno vrednostjo.
difference is also that after a decreased
application of atrazine the calculated values in
the percolate start decreasing quickly, whereas
the measured values decrease slowly. The
reason for that is that in the calculated values
only the layer at 10 m is considered, whereas
the measured values are affected by a wide
drainage basin from which the atrazine leaches
into the groundwater. A good example of this
is the Hrastje water pumping station, which
has the most extended drainage basin. That is
probably why concentrations in the water
pumping station are still above the permitted
values.
0
1
2
3
4
5
6
7
8
9
10
1
10
100
1000
Koncentracija / Concentration [ µg/l]
Globina / Depth [m]
13.05.1995
23.06.1995
13.01.1996
06.05.1996
Slika 4. Koncentracija atrazina v precedku v lahki in srednje težki zemljini
na razli čnih globinah ob razli čnih časih.
Figure 4. Concentration of atrazine in the percolate of light and moderate soil
at different depths and at different dates.

  Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
108
0
1
2
3
4
5
6
01.01.1989
01.01.1990
01.01.1991
01.01.1992
01.01.1993
01.01.1994
01.01.1995
01.01.1996
01.01.1997
01.01.1998
01.01.1999
01.01.2000
01.01.2001
01.01.2002
01.01.2003
01.01.2004
Koncentracija / Concentration   µg/l
Slika 5. Koncentracija atrazina v precedku v lahkih in srednje težkih tleh na globini 10 m.
Figure 5. Concentration of atrazine in percolate in light and moderate soil at a depth of 10 m.
0
0.05
0.1
0.15
0.2
0.25
01.01.1989
01.01.1990
01.01.1991
01.01.1992
01.01.1993
01.01.1994
01.01.1995
01.01.1996
01.01.1997
01.01.1998
01.01.1999
01.01.2000
01.01.2001
01.01.2002
01.01.2003
01.01.2004
Koncentracija / Concentration   µg/l
Slika 6. Koncentracija metolaklora v precedku v lahkih in srednje težkih tleh na globini 10 m.
 Figure 6. Concentration of metolaclor in percolate in light and moderate soil at a depth of 10 m.

  Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
109
Preglednica 12. Maksimalne koncentracije atrazina v precedku na globini 10 m.
Table 12. Maximum concentrations of atrazine in the percolate at 10 m depth.
Maksimalna
koncentracija
Maximum
concentration
Datum maksimalne
koncentracije
Date of maximum
concentration
Frakcija atrazina, ki je
pronicala v podtalnico
Fraction of atrazine
leached in groundwater
[ µg/l] [%]
Lahka tla
Light soil
18.51 18.11.1991 8.9
Lahka in srednje težka tla
Light and moderate soil
5.83 16.12.1991 3.2
Srednje težka tla
Moderate soil
1.98 06.11.1993 1.1
Železniški nasip
Railway dike
96.75 12.08.1991 67.8
Da bi se prepri čali o uporabnosti modela,
smo simulirali tudi distribucijo metolaklora po
zemljini. Metolaklor smo izbrali, ker se je
uporabljal v približno istih koli činah kot
atrazin, vendar pa ga v podtalnici niso zaznali.
Rezultati so potrdili, da je metolaklor veliko
manj mobilen kot atrazin, saj je model za
lahka tla izra čunal do stokrat manjše vrednosti
(slika 6) kot za atrazin. Vendar se te vrednosti
zelo hitro spremenijo (pove čajo ali zmanjšajo)
ob spremembi koeficientov, ki najbolj vplivajo
na mobilnost pesticidov. Ti koeficienti so
absorpcija na organsko snov (K
om
), razpolovni
čas in topnost v vodi. V literaturi smo za
atrazin našli koeficiente K
om
 v razli čnih
zemljinah, ki so se med seboj razlikovali tudi
za faktor 3. Pri atrazinu so izra čunane
koncentracije atrazina v lahkih tleh tako visoke
zaradi tega, ker je za zemljino, ki se nahaja na
lahkih tleh Ljubljanskega polja, koeficient K
om
najmanjši, kar pomeni, da se pesticid na tako
zemljino ne veže, ampak pronica v podtalje.
Za metolaklor pa takih podatkov nismo našli.
Morda je tudi to razlog za tolikšno razliko med
koncentracijama.
4. ZAKLJU ČKI
Slabost modela je ta, da ra čuna samo do
globine 10 m. Vseeno menimo, da je model
uporaben tudi za globlje podtalnice, saj se
izra čunana koncentracija atrazina v precedku
To check the applicability of the model we
then simulated the distribution of metolaclor in
the soil. Metolaclor was chosen because it was
used in the same amounts as atrazine, but it
was not found in the wells of the water
pumping stations. Results confirmed that
metolaclor is less mobile than atrazine. The
calculated values (Figure 6) of metolaclor in
percolate were up to a hundred times smaller
than those of atrazine. But these values can
change quickly (increase or decrease) by
changing the coefficients that mostly influence
the pesticide mobility. These coefficients
include sorption on organic matter (K
om
), half
life, and solubility in water. In the literature,
we found coefficients K
om
 for atrazine in
different soils; they vary up to a factor of 3.
The calculated atrazine concentrations in light
soil were found so high because coefficient
K
om
for such soil is the lowest, meaning that
the pesticides do not bound to the soil, but
percolate into the groundwater. For
metolaclor, no such data were found. Maybe
that is the reason for such a discrepancy in
percolate concentrations of atrazine and
metolaclor.
4. CONCLUSIONS
The weakness of the model is that the
maximum depth of the calculation is 10 m.
Nevertheless, we believe that the model is
suitable for deeper groundwater, because the
concentration in the percolate at a depth of 7.5

  Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
110
na globini 7,5 m skoraj ne razlikuje od
koncentracije na 10 m (glej sliko 4). Nastane
samo časoven zamik zaradi potovanja
pesticida po zemljini navzdol. Model se lahko
uporabi za prvo oceno obnašanja pesticidov v
tleh, preden se jih registrira in za čne
uporabljati. Kvaliteta simulacije modela je
odvisna od kvalitete vhodnih podatkov, ki jih
je treba dolo čiti z laboratorijskimi preiskavami
in terenskimi meritvemi.
Na Ljubljansko polje je bilo po ocenah od
leta 1991 do 1995 letno nanešeno 450 kg
atrazina. Dinami čne zaloge podtalnice se po
ocenah gibljejo med 3 in 4 m
3
/s, to je
126.144.000 m
3
/leto. Da se z atrazinom do
mejne vrednosti 0,1 µg/l onesnaži 126.144.000
m
3
 vode, je dovolj že 12,6 kg atrazina, kar je
2,8 % od 450 kg.
Z modelom PEARL smo izra čunali, da do
podtalnice skozi lahka tla pride 8,9 % atrazina,
skozi lahka in srednje težka tla 3,2 %, skozi
srednje težka tla 1,1 % atrazina in skozi
železniški nasip 67,8 % atrazina. V podtalju ni
intenzivnega mešanja vode, zato so bile
vrednosti atrazina v vodarnah višje, kot pa bi
bile v primeru popolnega premešanja. Iz tega
lahko sklepamo, da je model kar dobro
izra čunal odstotek izcejanja atrazina v
podtalnico.
Model PEARL omogo ča ocenjevanje
posledic uporabe razli čnih vrst pesticidov na
onesnaženje podtalnice in je kot tak dobro
orodje za oceno razli čnih ukrepov.
ZAHV ALA
Avtor se zahvaljuje prof. dr. Mitji Brillyju
in prof. dr. Branetu Mati či ču za strokovno
pomo č. Za podatke, ki jih je uporabil pri
pripravi članka, se zahvaljuje Kmetijskemu
inštitutu Slovenije in Agenciji RS za okolje.
m is almost identical to the concentration at 10
m (see Figure 4). The difference is only in
terms of time when the concentration reaches
the maximum. The reason for that is the travel
of a pesticide through soil. The model can be
used for a first evaluation of pesticide mobility
before they are registered and used. Quality of
simulation is depended upon the quality of
input data, which must be defined with
laboratory analysis and field measurements.
It has been evaluated that from 1991 to
1995, 450 kg of atrazine was applied yearly to
the fields of the Ljubljansko polje. According
to evaluations, dynamic groundwater storages
are in a range of 3–4 m
3
/s, which amounts to
126,144,000 m
3
/year. For contamination of
126,144,000 m
3
 of groundwater with atrazine
up to the boundary value, which is 0.1 µg/l,
12.6 kg of atrazine is enough, this being just
2.8 % of 450 kg.
With the PEARL model we calculated the
amounts of atrazine, which leach into the
groundwater. Through light soil, 8.9 % of the
whole application percolates into the
groundwater, through light and moderate soil
3.2 %, through moderate soil 1.1 %, and
through the railway dike 67.8 % of atrazine. In
the aquifer, intensive mixing of water is not
present, and because of that values of atrazine
in water plants were higher than they would be
in the case of absolute mixing. Therefore we
can assume that the model yielded quite a
good estimation of the atrazine leaching into
the groundwater.
ACKNOWLEDGMENTS
The author would like to thank Prof. Mitja
Brilly and Prof. Brane Mati či č for their
professional support. For the data used for the
preparation of the article the author would like
to thank the Agricultural Institute of Slovenia
and the Environmental Agency of the Republic
of Slovenia.

  Klemen či č, S.: Uporaba modela PEARL za simulacijo emisije pesticidov v podtalnico Ljubljanskega polja –
Usage of model PEARL for simulation of pesticide emission into the Ljubljansko polje groundwater
© Acta hydrotechnica 22/37 (2004), 95–111, Ljubljana
111
VIRI – REFERENCES
Andjelov et al. (2005). Podtalnica Ljubljanskega polja. Ljubljana, 251 p.
ARS PPD (2005). Agriculture Research Service Pesticide Properties Database.
http://www.ars.usda.gov/Services/docs.htm?docid=6433 (20. 1. 2004)
Leistra, M., van der Linden A. M. A., Boesten, J. J. T. I., Tiktak, A. & van der Berg, F. (2001).
PEARL model for pesticide behaviour and emissions in soil–plant systems – Descriptions of
the processes in FOCUS PEARL version 1.1.1. Research Report. Wageningen. RIVM,
ALTERRA Green World Research, Wageningen University and Research Center.
    http://www.alterra-research.nl/pls/portal30/docs/folder/pearl/pearl/pdf/pearlthe.pdf
(20. 1. 2004)
Simon či č, A., Zupan, M., Knapi č, M., Vrš čaj, B., Suhadolc, M., Šporar, M., Rupreht, J., Bukovec,
P. & Ti č, I. (2004). Izdelava meril za pripravo postopkov in programov uporabe
fitofarmacevtskih sredstev na obmo čjih z omejeno uporabo (Drawing-up of criteria for
preparation of procedures and programs concerning plant-protection products in areas of
limited use). Final Report. Žalec. Inštitut za hmeljarstvo, gozdarstvo in prehrano RS, Univerza
v Ljubljani, Biotehni čna fakulteta, Kmetijski inštitut Slovenije. (In Slovenian)
SWCHPC (2004). Soil Water Characteristics Hydraulic Properties Calculator. Washington DC,
USDA Agricultural Research Service.
Tiktak, A., van den Berg, F., Boesten, J. J. T. I., van Kraalingen, D., Leistra M. & van der Linden
A. M. A. (2000). Manual of FOCUS PEARL version 1.1.1. Research Report. Wageningen.
RIVM, ALTERRA Green World Research, Wageningen University and Research Center.
http://www.alterra-research.nl/pls/portal30/docs/folder/pearl/pearl/pdf/pearlman.pdf
(20. 1. 2004)
Naslov avtorja – Author's Address
Sašo Klemen či č
Univerza v Ljubljani – University of Ljubljana
Fakulteta za gradbeništvo in geodezijo – Faculty of Civil and Geodetic Engineering
Jamova 2, SI-1000 Ljubljana, Slovenia
E-mail: saso.klemencic@siol.net