© Acta hydrotechnica 20/33 (2002), Ljubljana
ISSN 0352-3551
371
UDK / UDC: 519.61/.64:504.4.054:551.468(72) Prejeto / Received:15.7.2002
Izvirni znanstveni prispevek – Original scientific paper Sprejeto / Accepted: 28.12.2002
MODELIRANJE ONESNAŽEVANJA MEHIŠKIH LAGUN S
POLJEDELSKIH POVRŠIN – PRVI DEL: METODOLOGIJA
MODELLING OF AGRICULTURAL POLLUTION IN MEXICAN LAGOONS
– PART 1: METHODOLOGY
Matjaž Č ETINA, Rudi RAJAR, Marina PINTAR, Fernando GONZÁLEZ-FARIAS
V številnih mehiških obalnih lagunah gojijo ribe, škampe in školjke. Več inoma pa so v zaledju lagun
poljedelska območ ja, s katerih z vodo odtekajo v lagune tudi hraniva in pesticidi, ki pogosto
onesnažujejo morsko hrano nad dovoljeno mejo. Cilj raziskave je predlagati nekaj optimalnih
nač inov poljedelskega upravljanja, ki bo v zadostni meri zmanjšalo onesnaženje vode in morske
hrane v lagunah. V prispevku opisujemo metodologijo reševanja problema s sistemom dveh
numerič nih modelov. Z modelom GLEAMS (Groundwater Loading Effects of Agricultural
Management Systems) najprej za znane podatke o padavinah, tipu tal, tipu pridelka in vrsti hraniv
in pesticidov določ imo iztok vode, sedimentov, hraniv in pesticidov z območ ja. Potem z modelom
PCFLOW2D izrač unamo širjenje in koncentracije onesnaževalcev v lagunah. S pomoč jo obeh
modelov in z nekaj nujnimi meritvami bomo predlagali optimalni nač in poljedelskega upravljanja.
Ta prispevek opisuje predlagano tehnologijo, v drugem delu bo podrobno opisano modeliranje
cirkulacije in širjenja polutantov za sistem lagun ob dolini Carizo v SZ Mehiki, v tretjem delu pa bo
opisana uporaba modela GLEAMS in predlagane konč ne rešitve.
Ključ ne besede: dvodimenzijsko hidrodinamič no modeliranje, modeliranje polutantov, model
GLEAMS, obalne lagune, Mehika, marikultura
Many coastal lagoons in Mexico are used for mariculture (fish, shrimps, oysters). Regions near
lagoons are usually exploited by agriculture, and drain water from this area brings nutrients and
pesticides into the lagoons, which can contaminate products of mariculture. The goal of the
research is to propose some scenarios of “optimum” agricultural management, which will enable
diminished contamination of the lagoon water and marine food. First, several possible scenarios of
improved agricultural management are studied. For each scenario numerical simulations are
carried out by two models. Simulations by the GLEAMS model (with input parameters of hydrology,
soil type and applied nutrients and pesticides) gives information on the contaminant contents in the
drain water from the fields. Further on, simulations by two-dimensional hydrodynamic and
pollutant transport model PCFLOW2D shows the distribution of contaminant concentration in the
lagoons. By means of both models and some measurements, an optimum scenario of agricultural
management will be proposed. This paper describes the proposed methodology, the second part will
describe in detail the hydrodynamic and pollutant transport simulations for the case study of the
Carizo valley in NW Mexico, and the third part will deal with modelling of agricultural pollution by
the GLEAMS model and proposed management for the same case study.
Key words: twodimensional hydrodynamic modelling, modelling of pollutants, GLEAMS model,
coastal lagoons, Mexico, mariculture

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
372
1. OPIS PROBLEMA
V Mehiki je več je število sistemov lagun,
kjer marikultura predstavlja pomembno
gospodarsko panogo. Glavni produkti so ribe,
školjke in škampi. Navadno pa se območ ja
okrog lagun uporabljajo za poljedelstvo. Na
območ ju mehiških držav Sonora in Sinaloa je
več kot 1 250 000 ha poljedelskih površin, ki
jih namakajo, vse pa ležijo v bližini 44 obalnih
lagun, kjer imajo velik ulov rib.
Problem je v dejstvu, da za izboljšanje
poljedelskih pridelkov uporabljajo gnojila in
pesticide. Voda, s katero namakajo, pa spira te
polutante v lagune, kjer pogosto onesnažujejo
morsko hrano. V glavnem imajo hranila vedno
škodljiv uč inek na kakovost vode v lagunah, v
večjih koncentracijah lahko povzroč ijo
evtrofikacijo. Pesticidi pa so toksič ni in
dopustne koncentracije v vodah, kjer se goji
marikultura so vedno zelo majhne.
Glavni cilj raziskave je predlagati
metodologijo, s pomoč jo katere bi zmanjšali
koncentracijo hraniv in pesticidov v lagunah
tako, da bi ostala v sprejemljivih mejah. To bi
dosegli z naslednjimi postopki.
Najprej je treba določ iti možne postopke za
zmanjšanje vnosa hraniv in posebno še
pesticidov s polj v lagune. Mogoč ih je več ukrepov, predvsem v obliki izboljšanega
poljedelskega upravljanja. V ta namen
uporabljamo model GLEAMS. Uč inek raznih
ukrepov potem simuliramo s hidrodinamič nim
modelom PCFLOW2D, rezultati so podani v
obliki č asovnega in prostorskega razporeda
koncentracije onesnaženja v lagunah. To
omogoča primerjavo učinka posameznih
ukrepov ter določitev optimalnih. Shema
metodologije uporabe dveh modelov za
simulacijo poljedelskega onesnaževanja
obalnih lagun je podana na sliki 1. Na koncu
bo narejena tudi ekonomska primerjava
posameznih ukrepov, kjer bomo upoštevali
tako agronomske kot tudi marikulturne vidike
ekonomskega vrednotenja.
1. PROBLEM DESCRIPTION
There are numerous coastal lagoons in
Mexico, where mariculture represents an
important economic source of income. The
main products are fish, shellfish and shrimps.
The regions around these coastal lagoons are
often exploited by a griculture. In the regions
of Sonora and Sinaloa there are more than
1 250 000 ha of agricultural area, which are
irrigated and are situated in the neighborhood
of 44 coastal lagoons with high fish
production.
The problem is that fertilizers and
pesticides are used to increase the agricultural
production, and the irrigation water from the
fields transports these contaminants into the
lagoons, where they can contaminate marine
food. Basically the nutrients, resulting from
different fertilizers, have a damaging effect on
the lagoon water quality, and in greater
concentrations they can cause eutrophication.
Pesticides are toxic and their allowed
concentration in water bodies where
mariculture is applied is very low.
The main goal of the research is to propose
a methodology for ways of diminishing the
concentration of nutrients and pesticides in the
lagoons and for keeping it below some
allowable limit. This will be achieved by the
following steps.
Firstly, it is necessary to determine possible
measures for diminishing the input of
fertilizers and especially pesticides from the
agricultural area to the lagoons. Several
possible remediation measures are studied,
mainly in the form of improved agricultural
management. The GLEAMS model is used for
this purpose. The effect of different measures
is further on simulated by the hydrodynamic
and pollutant transport numerical model
PCFLOW2D, and the results are given as
spatial and temporal distribution of
contaminant concentration inside the lagoons.
This makes it possible to compare the effect of
different remediation measures and to propose
the optimum measures. Scheme of the
modelling methodology for the simulation of
agricultural pollution of a system of coastal
lagoons is given in Figure 1. An economical
evaluation of all the possible measures will
also be carried out, where both agricultural and
maricultural production will be taken into
account.

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
373
V študiji so obravnavani tudi nekateri
možni “hidravlič ni” ukrepi, npr.:
1. Izgradnja cevovodov, po katerih bi speljali
onesnaženo vodo s polj neposredno v
ocean. Ta rešitev je vprašljiva z
ekonomskega vidika.
2. Zasaditev določ enih rastlin v nekaterih
delih lagun, ki odvzemajo hraniva iz vode,
bi lahko pripomogla k izboljšanju
kakovosti vode v lagunah.
Several possible “hydraulic” remediation
measures will also be studied, e.g.:
1. To construct pipelines, which would
conduct contaminated water from the fields
directly to the open ocean. This solution is
questionable from the economic point of
view.
2. To grow water plants in some parts of the
lagoons, which generate sufficient intake
of nutrients and pesticides from the
contaminated water, and thus improve
water quality in the lagoons.
Slika 1. Shema metodologije uporabe dveh modelov
za simulacijo poljedelskega onesnaževanja obalnih lagun.
Figure 1 . Scheme of the modelling methodology for the simulation
of agricultural pollution of a system of coastal lagoons.

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
374
Eden od lagunskih sistemov je v dolini
Carizo v SZ Mehiki, v državi Sinaloa: lagune
Bacorehuis, Jitzamuri, in Agiabampo. Celotna
površina lagun je okrog 150 km
2
 in v bližini je
46 000 ha poljedelskih površin, od koder se
voda izteka v lagune. Lagune so v glavnem
plitke, povpreč na globina lagune Bacorehuis je
5 m, Agiabampo 2 m, globina južnega dela
lagune Jitzamuri pa samo 1 m. Lagunski
sistem je povezan s Tihim oceanom (oziroma s
Kalifornijskim zalivom za polotokom Baja
California) z ožino, ki ima največ jo globino
13 m. Podrobnejši opis območja je v
nadaljevanju č lanka (Č etina et al., 2003).
Opravljena je bila predhodna študija
modeliranja hidrodinamične cirkulacije in
disperzije onesnaženja s teh poljedelskih
območ ij in je opisana v drugem delu tega
prispevka (Č etina et al., 2003).
2. MODELIRANJE IZTOKA
ONESNAŽEV ALCEV S
POLJEDELSKIH OBMOČ IJ:
MODEL G L E A M S
2.1 POLJEDELSKO UPRAVLJANJE
V svetu je dušik (N) glavni omejitveni
dejavnik za celotno proizvodnjo hrane kot tudi
za vsebnost proteina (Peterson & Frye, 1989),
zato je pravilno upravljanje z dušikom izredno
pomembno tako za proizvodnjo hrane, kot tudi
za onesnaževanje okolja. V primeru obalnih
lagun, ostanki namakalne vode, ki odteka s
polj s seboj prinaša onesnaževalce in povzroč a
omenjeno onesnaževanje morske hrane.
Zato je pravilno upravljanje na poljih zelo
pomembno za zmanjšanje onesnaževanja s
hranivi in pesticidi. Tu gre predvsem za
pravilno izbiro tipa hraniva, količ ine
uporabljenega hraniva, metode aplikacije,
pravilno izbiro č asa gnojenja, vrste in količ ine
pesticida, metode aplikacije pesticida ter
pravilno izbira č asa uporabe pesticidov.
One of the lagoon systems is situated in the
Carizo valley in NW Mexico, in the Mexican
state of Sinaloa: the lagoons Bacorehuis,
Jitzamuri, and Agiabampo. The total area of
the lagoon system is about 150 km
2
 and there
is 46 000 ha of agricultural area which is
drained into the lagoons. The lagoons are
mainly shallow, the average depth of
Bacorehuis being 5 m, of Agiabampo 2 m, and
the depth of the southern part of Jitzamuri only
1 m. The lagoon system is connected with the
Pacific Ocean (in fact with the California Bay
behind the Baja California peninsula) b y a
strait, with a maximum depth of 13 m. A more
detailed description of the region is given in
Č etina et al. 2003.
A preliminary study of modelling of the
transport–dispersion of contaminants from
these agricultural regions has been carried out
and is described in greater detail in the second
part of this article (Č etina et al., 2003).
2. MODELLING OF OUTFLOW OF
CONTAMINANTS FROM FIELDS:
THE G L E A M S MODEL
2.1 AGRICULTURAL MANAGEMENT
Since nitrogen (N) fertilizer supply is the
primary nutrient limitation for both total world
food production and protein content (Peterson
& Frye, 1989), effective management of N is
extremely important on the one hand for food
production and, on the other hand, also for
preventing contamination of the environment.
In the case of coastal lagoons, the irrigation
water brings contaminants into the lagoons,
thus causing contamination of marine food.
Therefore proper agricultural management
is of great importance for diminishing water
contamination by fertilizers and/or pesticides.
This involves the following decisions
regarding the proper: type of applied fertilizer;
amount of fertilizer; method of application;
timing of fertilizer application; type of applied
pesticides; amount of pesticide; method of
pesticide application; timing of pesticide
application.

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
375
Iz zgornjega je razvidno, da je prvi in
pogosto najbolj uč inkovit ukrep za zmanjšanje
onesnaževanja vode v obalnih lagunah
optimalno poljedelsko upravljanje. Ta
metodologija bo bolj natanč no opisana v
tretjem delu prispevka (Pintar et al., 2003).
Shema modela GLEAMS je podana na sliki 2.
From the above it can be understood, that
the first, and often the most effective step in
solving the problem of lagoon contamination
is to determine optimum agricultural
management. This methodology will be
described more in detail in the third part of this
paper (Pintar et al., 2003). A scheme of
GLEAMS model is given in Figure 2.
Vhodni podatki
Rezultati
Lastnosti tal
Topografija
Uporaba
Obdelava tal
Hraniva
Pesticidi
Padavine
Globalno
sevanje
Temperatura
Veter
Evapotranspiracija
Površinski odtok
Perkolacija
Adsorbirani in absorbirani
polutanti
Slika 2. Shema modela GLEAMS.

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
376
Input data
Output data
Soil properties
Topography
Land use
Tillage system
Nutrients
Pesticides
Precipitation
Global
radiation
Temperature
Wind
Evapotranspiration
Surface runoff
Percolation
Adsorbed and
dissolved pollutants
Figure 2. Scheme of GLEAMS model.
2.2. VPLIV KMETIJSTVA – UPORABA
MODELA GLEAMS
Naslednji del raziskave mora dati odgovor
na vprašanje: “Koliko mineralnih gnojil in
fitofarmacevtskih sredstev se spere v lagune z
opazovanega kmetijskega območja pri
različnih scenarijih kmetijske prakse?”
Vprašanje je pomembno za določ itev vpliva
posameznega scenarija in določ itev
najprimernejše kmetijske prakse. Pri tem se
uporablja matematično modeliranje s
posameznimi meritvami.
Eden najbolj uporabnih modelov, ki
simulirajo omenjeni fenomen, je GLEAMS
(Groundwater Loading Effects of Agricultural
Management Systems, Obremenitve podtalnice
ob upravljanju s poljedelskimi sistemi), ver.
2.2 DETERMINATION OF THE EFFECT
OF AGRICULTURAL MANAGEMENT –
USE OF GLEAMS MODEL
The next part of the research must give an
answer to the question: “What amounts of
fertilizers and pesticides are being washed
from the studied agricultural area into lagoons
at each scenario of agricultural management?”.
This question is important in order to
determine the effect of each scenario and then
to find out the most appropriate management.
Numerical modelling will be used for this
purpose, together with some measurements.
One of the most appropriate models for
simulating this phenomenon is the GLEAMS-
Groundwater Loading Effects of Agricultural
Management Systems, ver. 3.0. This is a
mathematical simulation model developed for

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
377
3.0. Ta matematič ni simulacijski model so
razvili za območja velikosti polja za
ovrednotenje vplivov kmetijske prakse na
gibanje snovi znotraj in skozi koreninsko talno
plast. Model uporabljajo v več kot 40 državah,
na voljo pa je tudi prek interneta.
Glavni vhodni podatki modela GLEAMS
so:
− hidrološki podatki območ ja (dnevni podatki
o padavinah in temperaturi; meseč ni
podatki o temperaturi, sonč no sevanje, veter
in temperatura rosišč a),
− lastnosti tal,
− podatki o kmetijski kulturi,
− znač ilnosti fitofarmacevtskega sredstva in
podatki o aplikaciji in
− gnojenje in podatki o obdelavi tal.
Rezultati modela so dnevni, meseč ni in letni
podatki:
− vodni tok (površinski ali podpovršinski ali
oboje,
− premešč anje sedimentov,
− izgube hranil (količ ina in koncentracija),
− izgube fitofarmacevtskih sredstev (količ ina
in koncentracija).
Kniesel (1995), ki je eden od avtorjev
modela GLEAMS, je tega na kratko opisal:
Model GLEAMS je sestavljen iz štirih
komponent, ki potekajo soč asno: hidrologija,
erozija/nanos materiala, fitofarmacevtska
sredstva in rastlinska hranila.
Hidrologija. Simulirana je dnevna vodna
bilanca za talni sistem v koreninski coni.
Model razporedi talne znač ilnosti na največ 12
rač unskih plasti, pri tem imajo tla do pet talnih
horizontov. Izrač unana je dnevna potencialna
evapotranspiracija, odtok, dostopna količ ina
vode v tleh, odtok skozi talni profil in
namakanje.
Erozija. Komponenta erozije je
modifikacija Universal Soil Loss Equation
(USLE). Faktor erozivnosti (R) iz omenjene
enačbe nadomesti energija dežja v
posameznem deževnem dogodku. Premešč anje
sedimentov se rač una na podlagi specifič nega
pretoka, ki ga v hidrogeološki komponenti
računamo iz simuliranega največ jega
površinskega odtoka za nevihto na iztoku z
območ ja.
field-size areas to evaluate the effects of
agricultural management systems on the
movement of agricultural chemicals within
and through the plant root zone. The model is
widely used in over 40 countries and is
available on the Internet.
The basic input data for the GLEAMS
model are:
− Hydrologic data of the region (daily rainfall
and temperature; monthly temperature,
solar radiation, wind movement, and
dewpoint temperature),
− Soil characteristics
− Crop situation,
− Pesticide characteristics and application
data, and
− fertilization and tillage data.
The model output results are various-daily,
monthly, annual:
− Water flow (either surface or groundwater
or both),
− Sediment transport,
− Nutrient outflow (mass and concentration),
− Outflow of pesticides (mass and
concentration).
Kniesel (1995) is one of the authors of the
GLEAMS model. He briefly describes the
model:
The GLEAMS model consists of four
components operating simultaneously:
hydrology, erosion/sediment yield, pesticides,
and plant nutrients.
Hydrology. Daily water accounting is
simulated in a soil system layered within the
soil horizons of the root zone. The model
distributes soil characteristics into a maximum
of 12 computational layers with input from a
maximum of 5 soil horizons. Daily potential
evapotranspiration is calculated, as well as
runoff, available water storage, percolation
through the soil layers, and irrigation.
Erosion. The erosion component is the
modification of the Universal Soil Loss
Equation (USLE). The erosivity (R) factor of
the USLE is replaced by storm-by-storm
rainfall energy calculated from daily rainfall.
A characteristic discharge, calculated from the
storm runoff peak rate simulated at the field
outlet in the hydrogeology component, is used
to calculate sediment transport.

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
378
Fitofarmacevtska sredstva. Komponenta
fitofarmacevtskih sredstev vključ uje
površinski tok, kot je določ en v modelu
CREAMS z vertikalnim tokom ter definira pot
fitofarmacevtskih sredstev v, znotraj ter skozi
koreninsko cono. Adsorpcija fito-
farmacevtskega sredstva na talni organski
ogljik se uporablja za razdelitev komponente
med talno raztopino in tlemi za simulacijo
izluževanja v površinski odtok, sediment in
perkolirano vodo. Razgradni produkti
(metaboliti) se upoštevajo, kadar so znane
njihove znač ilnosti. Soč asno se lahko simulira
10 fitofarmacevtskih sredstev z razgradnimi
produkti. Simulacija izgub fitofarmacevtskih
sredstev poteka v površinskem odtoku,
sedimentu in perkolatu na spodnjem robu
koreninske cone.
Rastlinska hranila. Pri rastlinskih hranilih
sta upoštevana dušik in fosfor. V modelu sta
obravnavana njuna obsežna ciklusa. Aplikacija
anorganskih gnojil upošteva površinsko
aplikacijo, inkorporacijo v tla in fertigacijo.
Živinski odpadki, s podatkom o vsebnosti
hranil, so lahko aplicirani površinsko, lahko so
inkorporirani ali injektirani v tla, lahko so
porabljeni kot iztoki iz lagun. Algoritmi
obdelave in temperature tal so vključ eni v
komponento hranil. Dušik v padavinah je
vhodni podatek, vezan na območ je aplikacije
modela. Kjer so koncentracije v vodnem viru
več je, se v vodi za namakanje dušik in fosfor
upoštevata.
Kmetijska praksa. GLEAMS je bil razvit za
primerjavo odzivov različ nih kmeijskih praks
na podnebje. Model mora biti sposoben za
predstavitev širokega izbora kmetijskih praks.
Dobri in slabi hidrološki pogoji kategorizirajo
sposobnost tal za infiltracijo dežja. Metoda
torej predstavlja kmetijsko prakso s simulacijo
kontinuiranega odtoka.
S pomoč jo GLEAMSa lahko naredimo
izbor fitofarmacevtskih sredstev za specifič na
tla in kolobar, glede na znač ilnosti
fitofarmacevtskega sredstva. Če lahko
ustrezno kontrolo dosežemo z alternativnim
fitofarmacevtskih sredstvom, izberemo v smeri
zmanjšane izgube preko površinskega toka ter
perkolacije.
Pesticides. The pesticide component
incorporates the surface pesticide flux
response to CREAMS with a vertical flux
component to route pesticides into, within, and
through the root zone. Characteristics of
pesticide adsorption to soil organic carbon are
used to partition compounds between solution
and soil fractions for simulating extraction into
runoff, sediment, and percolation losses.
Degradation products, (metabolites) can be
considered when their characteristics are
known. Up to 10 pesticides can be simulated
simultaneously, including metabolites.
Pesticide losses are simulated in runoff, with
sediment, and in percolate at the bottom of the
root zone.
Plant nutrients. The plant nutrient
component considers nitrogen and phosphorus.
Comprehensive nitrogen and phosphorus
cycles are treated in the model. Inorganic
fertilizer application considers surface
application, incorporation, and fertigation.
Animal waste application, with specification
of nutrient content, may be represented as
surface, incorporation, injection, or liquid such
as lagoon effluent. Tillage and soil
temperature algorithms are included in the
nutrient component. Rainfall nitrogen is the
input for the model application site, and N and
P in irrigation water can be considered for
locations where concentrations in the water
supply may be significant.
Management. GLEAMS was developed to
compare the responses of alternate
management practices to long-term climate.
The model must be capable of representing a
wide range of management practices. Good
and poor hydrologic conditions have further
categorized receptiveness of the soil for
rainfall infiltration. Thus, the method
represents management in the continuous
runoff simulation procedure.
A selection of pesticides for specific soil
and cropping systems can be made with the
GLEAMS model based upon pesticide
characteristics. If adequate control can be
achieved with alternate pesticides, selection
can be made to minimize runoff losses or to
minimize percolation losses.

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
379
Pri hidrologiji se upošteva posamezna
kultura, vendar pa se lahko upošteva tudi
sestavljen indeks listne površine. Kolobar, ki
vljuč uje več kultur, predstavlja intenzivno
kmetijsko prakso, ki jo lahko upoštevamo v
GLEAMS-u.
GLEAMS se uporablja v več kot 40
državah v Severni Ameriki ter Evropi.
Njegovo uporabo priporoč a tudi American
Environmental Protection Agency. Dobri
rezultati validacije programa so bili doseženi
le v primerih, ko so bili uporabljeni lokalni
parametri, kar priporoč ajo tudi avtorji modela
(Wu et al., 1996; 1997). Uporabljen je tudi za
definiranje strategije zašč ite podtalnice (Diebel
et al., 1992). V slovenskih razmerah je bil
GLEAMS uporabljen za spremljanje atrazina v
tleh in je dal zadovoljive rezultate (Turk,
1995).
Vhodni podatki za model GLEAMS so
odvisni od scenarija kmetijske prakse. Kot
rezultat model pokaže vpliv kmetijske prakse,
zato je model GLEAMS uporaben za določ itev
prvih učinkov kmetijske prakse: iztok
onesnaževal s kmetijske površine v lagune.
2.3 MERITVE
Za umerjanje in preverjanje modela
GLEAMS morajo biti izvedene nekatere
meritve (od katerih, jih je del že opravljen):
− Tla: globina, tekstura, gostota tal,
poroznost, poljska kapaciteta tal za vodo,
toč ka venenja, vsebnost organske mase, pH,
nasič enost z bazami, vsebnost kalcijevega
karbonata. Vzroci tal bodo vzeti z določ ene
globine tal.
− Površina: oblika, naklon
− Podnebje: dnevne padavine, dnevne
temperature. Ostali podnebni podatki so
manj obč utljivi in so lahko pridobljeni in
obstoječ ih nizov podatkov za območ je.
Naslednji parametri bodo merjeni na iztoku
vode s kmetijske površine:
− površinski odtok vode,
− perkolacija,
− količ ina sedimenta,
A single crop is assumed in hydrology, but
a composite leaf area index can be input. Crop
rotation including multiple cropping, that is
growing more than one crop during a year,
represents intensive management that can be
considered in GLEAMS.
GLEAMS is used in over 40 North
American and European countries. Its use is
also recommended by the American
Environmental Protection Agency. Good
results for program validation were obtained
only in those cases where local parameters
were used, which is also recommended by the
program's authors (Wu et al., 1996; 1997). It is
used to define the ground water protection
strategy, as well (Diebel et al., 1992). In the
Slovene environment, GLEAMS has been
used to monitor atrazine in soil and yielded
satisfactory results (Turk, 1995).
The input data for the GLEAMS model
depend on the scenario of the agricultural
management. Since the model output results
show the effect of the management, the
GLEAMS model can be used to determine the
first effect of the management: the output of
contaminants from agricultural area into the
lagoons.
2.3 MEASUREMENTS
To calibrate and verify the GLEAMS model
several measurements should be carried out,
(some of them already done):
− Soil: soil depth, soil texture, soil bulk
density, soil porosity, field water capacity,
wilting point, content of organic mater in
soil, pH, base saturation, calcium carbonate
content. Soil samples will be taken to a
certain depth.
− Area: shape, inclination
− Climate: daily rainfall, daily temperature.
Other climate data are less sensitive and
should be gathered from existing sets of
climate data for the region.
The following parameters will be measured
in the outflow water from the agricultural area:
− water runoff,
− water percolation,
− sediment yield,

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
380
− uporabljena fitofarmacevtska sredstva bodo
merjena v površinskem odtoku, sedimentu,
v tleh in v perkolirani vodi,
− nitratni-N v površinskem odtoku, tleh in
perkolirani vodi,
− amonijski-N v površinskem odtoku,
sedimentu in tleh in
− fosfor v površinskem odtoku, sedimentu in
tleh.
Vodni vzorci bodo vzeti na iztoku
površinskih in/ali podzemnih drenov.
V letu 1998 je bila opravljena serija meritev
(González-Farias, 1998). Merjeni so bili
naslednji parametri: padavine, koncentracija
skupnega fosforja in skupnega dušika v
sedimentu drenov, lagun in v Oceanu.
3. MODELIRANJE CIRKULACIJE IN
DISPERZIJE KONTAMINANTOV V
LAGUNAH
3.1 PCFLOW2D MODEL
Naslednji korak raziskave je določ iti
krajevni in č asovni razpored koncentracije
kontaminantov v lagunah, kjer upoštevamo vse
opisane ukrepe za izboljšanje stanja. Določ ili
bomo tudi, č e je v katerikoli lokaciji presežena
dopustna koncentracija. Tu smo uporabili dvo-
dimenzionalni (2D) model PCFLOW2D. Ta
lahko v tem primeru nadomesti bolj
kompleksno 3D verzijo modela, saj so lagune
zelo plitve (globine so več inoma pod 5 m) v
primerjavi s horizontalnimi dimenzijami (več deset km) in več inoma ne prič akujemo več je
toplotne ali slanostne stratifikacije.
− pesticides which will be used on the
observed field, will be measured in runoff,
sediment, soil and percolated water,
− nitrate-N in runoff, soil and percolated
water,
− ammonia-N in runoff, sediment and soil,
and
− phosphorus in runoff, sediment and soil.
Water samples must be taken at the mouths
of the surface and/or underground drains.
In 1998, a measuring campaign was carried
out (González-Farias, 1998). The following
parameters have been measured: precipitation,
concentration of total phosphorus and total
nitrogen in the bottom sediment of the drains,
lagoons and in the Ocean.
3. MODELLING OF CIRCULATION
AND DISPERSION OF
CONTAMINANTS IN THE
LAGOONS
3.1 PCFLOW2D MODEL
The next step of the research is to determine
the spatial and temporal distribution of
contaminant concentration inside the lagoons,
taking into account all the described
agricultural and hydraulic remediation
measures. This will give information as to
whether at any location the allowable limit is
exceeded. This is carried out using a two-
dimensional numerical model PCFLOW2D.
This can replace here the more complex 3D
version, as the lagoons are very shallow (the
depth is mostly less than 5 metres) in
comparison to the horizontal dimensions
(several tens of kilometers) and mostly no
significant thermal or salinity stratification is
expected.

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
381
      NUMERIČ NI MODEL
             PCFLOW2D
HIDRODINAMIČ NI MODUL
 2D globinsko-povpreč ni
nelinearni
Metoda konč nih volumnov k-eps model turbulence
Hibridna shema
CONDIF
VZROKI GIBANJA:
veter, plimovanje, vtok rek 
TRANSPORTNO-DISPERZIJSKION  MODUL
  in  SEDIMENTATACIJSKI  MODUL
Met. konč nih vol. + CONDIF
Metoda sledenja delcev
Slika 3. Shema numerič nega modela PCFLOW2D.

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
382
      NUMERICAL MODEL
           PCFLOW2D
HYDRODYNAMIC MODULE
 2D depth-averaged
nonlinear
Finite volume Dif. Method k-eps turbulence model
Hybrid Scheme (Central/Upwind)
CONDIF
FORCING FACTORS:
wind,  tide,  river inflow
TRANSPORT-DISPERSION MODULE and
   SEDIMENTATION MODULE
Finite volume + CONDIF
Particle Tracking Method
Figure 3. Scheme of the numerical model PCFLOW2D.
Model je bil razvit na univerzi v Ljubljani
za simulacijo 2D tokov v rekah – z uporabo
samo hidrodinamič nega modula (glej sliko 3).
Kasneje je bil razvit transportno–disperzijski
modul za simulacijo širjenja nekaterih
polutantov v površinskih vodah. Ker je bil
model že več krat podrobno opisan (Rajar,
1992; Rajar & Č etina 1997), tu podajamo le
najosnovnejši opis.
Hidrodinamič ni modul. V tem delu se
rešuje kontinuitetna enač ba in 2D, globinsko
povpreč ne dinamič ne enač be. Uporabljamo
metodo konč nih prostornin s hibridno shemo
(kombinacija centralne in 'upwind' sheme).
Lahko upoštevamo naslednje robne pogoje
(R.P.) v poljubnih celicah rač unskega
The model was first developed at the
University of Ljubljana to simulate 2D flow in
rivers – using only a hydrodynamic module
(see Figure 3). Further on the transport-
dispersion module was developed to simulate
transport and mixing of different contaminants
in surface waters. As the model has been
described in detail in several papers (Rajar,
1992; Rajar & Č etina 1997), here only a brief
description is given.
Hydrodynamic (HD) module. It solves the
continuity and the 2D depth integrated
momentum equations. It is based on the finite
volume numerical method of solution with the
hybrid (central/upwind) solution scheme. The
following boundary conditions (B.C.) can be
used at arbitrarily chosen cells of the
computational domain: a) Solid boundaries

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
383
področ ja: a) trdne stene (normalne hitrosti so
nič ); b) vtok rek (predpisane so hitrosti ali
pretoki); c) podajamo globine ali kote gladine;
d) kritič ni tok; e) enač ba preliva. Vzdolž
odprtih robov so mogoč i naslednji robni
pogoji: a) znana krivulja plimovanja ali druga
podana č asovna funkcija spreminjanja gladine;
b) “radiacijski” robni pogoj; c) “kontinuitetni”
robni pogoj. Vpliv vetra je mogoč e upoštevati
skupaj s katerim koli drugim robnim pogojem,
tako da podamo hitrost vetra in trenjski
koeficient.
Transportno-disperzijski modul temelji na
advekcijsko-difuzijski enačbi in simulira
transport in disperzijo konzervativnih
onesnaževalcev. Za nekatere specifič ne
polutante (npr. nafta ali živo srebro), so
vključ eni nekateri biokemič ni procesi. Zaradi
uporabe hibridne numerič ne sheme je prisotna
določ ena numerič na difuzija, ki lahko v
določ enih pogojih povzroč i, da je toč nost
rezultatov vprašljiva. Zato smo razvili drugo
različ ico modula, ki temelji na Lagrange-
jevski metodi sledenja delcev (MSD). Ta
metoda nima numerič ne difuzije in jo je
mogoč e uč inkovito uporabiti za simulacijo
disperzije polutantov iz toč kovnih virov. V
primeru tega projekta smo uporabili metodo
MSD.
3.2 METODOLOGIJA
Slika 3 prikazuje temeljno metodologijo in
povezavo modelov GLEAMS in PCFLOW2D.
Lahko vidimo, da so izhodni rezultati modela
GLEAMS, to je iztok vode, sedimentov,
hraniv in pesticidov s poljskih površin,
istočasno vhodni podatki za model
PCFLOW2D. Ta model rač una
hidrodinamič no cirkulacijo vode v lagunah.
Največ krat na cirkulacijo v največ ji meri
vpliva plimovanje, vendar upoštevamo tudi
vpliv vetra in vtoka rek. Potem s transportno-
disperzijskim modulom simuliramo širjenje
onesnaževalcev v vodi. Konč ni cilj raziskave
je dobiti prostorski in časovni razpored
koncentracije hraniv in pesticidov v lagunah za
vsak predpostavljeni scenarij. Primerjava
razporeda koncentracije za predpostavljene
(with zero normal velocities); b) Inflow of
rivers (discharges or velocities can be
prescribed); c) Water depths or water surface
elevations; d) Critical flow; e) E quation of a
weir. Along the open boundaries the following
boundary conditions can be used: a) known
tidal or other time–dependent functions of the
water surface; b) “radiation” boundary
condition; c) “continuity” boundary condition.
Wind stress can be taken into account with all
boundary conditions by giving the wind speed
and the wind friction coefficient.
Transport-dispersion (TD) module is based
on the advection-diffusion equation and
simulates transport and dispersion of any
conservative contaminant. For some specific
contaminants (eg. oil or mercury), bio-
chemical processes are also included. Due to
the application of the hybrid (central–upwind)
scheme, a certain amount of numerical
diffusion is present, which can in some cases
render the results unreliable. Therefore another
version of this module was developed, which
is based on the Lagrangian Particle Tracking
Method (PTM). This method is free of
numerical diffusion and can be especially
effectively used for simulation of pollutant
dispersion from point sources. For the
simulations of contaminant dispersion in the
lagoons of Carizo valley this PTM method was
used.
3.2 METHODOLOGY
Figure 3 shows the basic methodology, and
linking of models GLEAMS and PCFLOW2D.
As can be seen, the output results of GLEAMS
i.e. information on water discharge, sediments,
nutrient and pesticide concentration in the
drain water from the fields are the input data
for PCFLOW2D. Running this model, first the
HD module calculates the hydrodynamic
circulation of water in the lagoons. Usually the
most important forcing factor, causing the
water circulation in coastal lagoons is tide, but
the influence of wind and river inflow is also
taken into account. Further on, the transport-
dispersion module simulates the transport and
mixing of contaminants in the water. The final
goal of the research is to obtain the spatial and
temporal distribution of nutrient and pesticides
concentration in the lagoons for each scenario.
Comparing the resulting contaminant
distribution for each scenario indicates the

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
384
scenarije pokaže optimalne rešitve.
Kot je že omenjeno, so že izvedene
simulacije cirkulacije v lagunah ob dolini
Carizo opisane v č lanku Č etina et al. (2003).
3.3 MERITVE
Za umerjanje in verifikacijo modela
PCFLOW2D bomo izvedli v lagunskem
sistemu doline Carizo naslednje meritve (da bi
postopek ekonomizirali, bo število meritev
reducirano na minimum in bomo uporabljali
preproste metode meritev, kjerkoli je mogoč e):
− Meritve gladin kot funkcija č asa v 2 do 3
lokacijah v lagunah.
− Meritve hitrosti toka v 2 do 3 lokacijah v
lagunah.
− Meritve disperzije sledila v lagunah.
Merjeni potek gladin bomo primerjali s
simuliranimi. Spremembe gladine več inoma
povzroč a plimovanje na odprtem robu, manjši
del sprememb gladine povzroč i tudi veter,
katerega vpliv je tudi upoštevan v modelu. Za
umerjanje in verifikacijo transportno–
disperzijskega modula, bomo merili č asovni
potek disperzije sledila. Uporabljali bomo
razgradljivo sledilo, ki ne onesnažuje okolja.
4. ZAKLJUČ KI
V tem delu č lanka je opisana metodologija
reševanja. Cilj je doseč i izboljšano poljedelsko
upravljanje, oziroma zmanjšati vnos hraniv in
pesticidov v lagune. Simulacije z modeloma
GLEAMS in PCFLOW2D bodo pomagale
določ iti optimalne možne ukrepe.
Drugi del č lanka (Č etina et al., 2003)
opisuje simulacije hidrodinamič ne cirkulacije
in disperzije onesnaževalcev v lagunah doline
Carizo. V tretjem delu (Pintar et al., 2003) bo
opisana uporaba modela GLEAMS za
simulacijo različnih metod poljedelskega
upravljanja v sistemu doline Carizo v SZ
Mehiki.
optimum solutions.
As already mentioned, several simulations
of the circulation and contaminant dispersion
in the coastal lagoons of the Carizo valley are
presented in Č etina et al. (2003).
3.3 MEASUREMENTS
To calibrate and verify the model
PCFLOW2D, the following measurements
will be carried out in the lagoons (in order to
economize, these measurements will be
reduced to the minimum, and simple
measuring methods will be used, whenever
possible):
− Measurements of water level elevations as a
function of time at two to three locations
inside the lagoons.
− Measurements of velocity at two to three
locations in the lagoons.
− Measurements of tracer dispersion in the
lagoons.
Measured water level elevations will be
compared with the simulated elevations.
Mainly the water level rise is caused by the
tidal fluctuation, a small part of the elevation
may be due to wind forcing, which is also
taken into account in the model. To calibrate
and verify the transport–dispersion part of the
model, a tracer will be put into the lagoon and
the time development of its dispersion will be
measured. A degradable tracer will be used,
with no harmful effects for the environment.
4. CONCLUSIONS
The methodology to be used to solve this
problem is described in this part of the paper.
Improved agricultural management will be
studied, the effect will be simulated by two
models: GLEAMS and PCFLOW2D.
Simulations will help determine the best
solution.
The second part of the paper (Č etina et al.,
2003) describes hydrodynamic and
contaminant transport simulations, and the
third part (Pintar et al., 2003) will describe the
use of the GLEAMS model to simulate several
possible agricultural management
methodologies in the system of the Carizo
valley, NW Mexico.

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
385
VIRI – REFERENCES
Boisvert, R. N., Regmi, A., Schmit, T. M. (1997). Policy Implications of Ranking Distributions of
Nitrate Runoff and Leaching From Corn Production by Region and Soil Productivity. Journal of
productional agriculture 10, 102–108.
Č etina, M., Rajar, R., Pintar, M., González–Farias, F. (2003). Modelling of agricultural pollution in
Mexican lagoons – Part 2: Hydrodynamic and pollutant dispersion modelling for Case Study –
Carizo Valley. (submitted to Acta hydrotechnica).
Diebel, P. L., Taylor, D. B., Batie, S. S., Heatwole, C. D. (1992). Low–input Agriculture as a
Ground Water Protection Strategy. Water Resources Bulletin 28, 755–761.
González–Farias, F., Cisneros Estrada, X., Fuentes Ruíz, C. (1998). Evaluación del impacto
ambiental causado por agroquímicos en lagunas costeras del noroeste de México. Informe Final
Proyecto Clave: RD–9825 CAN–IMTA. 34 p.
Peterson, G.A., Frye, W.W. (1989). “Fertilizer Nitrogen Management” in R.F. Follet, Ed., Nitrogen
Management and Ground Water Protection. Series: Development in Agricultural and Managed–
Forest Ecology 21, Elsevier.
Pintar, M., Pikl, M., Lobnik, F., Hudnik, V. (1996). Nitrate and Atrazine Transfer From Rural
Catchment to Water – A Case Study in the Apač e Valley, Slovenia. Fresenius environmental
Bulletin 5, 649–654.
Pintar, M., Č etina, M., Rajar, R., González–Farias, F. (2003). Modelling of agricultural pollution in
Mexican lagoons – Part 3: Modelling of agricultural management (GLEAMS model) for Case
Study – Carizo Valley. (submitted to Acta hydrotechnica).
Pintar, M., Knapič , M. (1998). Nitrate Leaching in Hop Production. Fresenius environmental
Bulletin 7, 590–595.
Rajar, R. (1992). Application of the Three-Dimensional Model to Slovenian Coastal Sea.
International Conference on Computer Modelling of Seas and Coastal Regions, Southampton,
413–424.
Rajar, R., Č etina, M. (1997). Hydrodynamic and Water Quality Modelling: An Experience.
Ecological Modelling 101, 195–207.
Rajar, R., Č etina, M., Širca, A.(1997). Hydrodynamic and Water Quality Modelling: Case Studies.
Ecological Modelling 101, 209–228.
Rajar, R., Brebbia, C.A. (eds.) (1997). Water Pollution IV - Modelling, Measuring and Prediction.
Computational Mechanics Publication, Southampton (UK), Boston (USA), 811 p.
Rice, A. J., Viste, D. A. (1994). Major Sources of Groundwater Contamination: Assesesing the
Extent of Point and Nonpoint Contamination in a Shallow Aquifer System.Groundwater
Contamination and Control, New York, Marcel Dekker, 21–36.
Turk, I. (1995). Applicability of Dynamic Model GLEAMS for Prediction of the Migration of
Atrazine in Soil in Slovenian Area. Unpublished Master’s thesis, University of Ljubljana,
Biotechnical Faculty, Dept. of Agriculture, 108 p.

Č etina, M., Rajar, R., Pintar, M., González-Farias F.: Modeliranje onesnaževanja mehiških lagun s poljedelskih površin
- prvi del: metodologija - Modelling of Agricultural Pollution in Mexican Lagoons - Part 1: Methodology
© Acta hydrotechnica 20/33 (2002), 371-386, Ljubljana
386
Naslovi avtorjev – Authors’ Addresses
izr. prof. dr. Matjaž Č etina
Univerza v Ljubljani – University of Ljubljana
Fakulteta za gradbeništvo in geodezijo – Faculty of Civil and Geodetic Engineering
Jamova 2, SI–1000 Ljubljana
E–mail: mcetina@fgg.uni-lj.si
prof. dr. Rudi Rajar
Orle 44, SI–1291 Škofljica, Slovenia
E–mail: rrajar@fgg.uni-lj.si
doc. dr. Marina Pintar
Univerza v Ljubljani – University of Ljubljana
Biotehniška fakulteta – Biotechnical Faculty
 Jamnikarjeva 101, SI–1000 Ljubljana
E–mail: marina.pintar@bf.uni-lj.si
Dr. Fernando González–Farias
Mexican Institute for Water Technology
Cuernavaca, Mexico
E–mail: gfarias@servidor.unam.mx