Available via license: CC BY-NC-ND 3.0
Content may be subject to copyright.
Using water and Agrochemicals in the Soil, Crop and
Vadose Environment (WAVE) model to interpret
nitrogen balance and soil water reserve under different
tillage managements
N. Zare, M. Khaledian, J.C. Mailhol
To cite this version:
N. Zare, M. Khaledian, J.C. Mailhol. Using water and Agrochemicals in the Soil, Crop and
Vadose Environment (WAVE) model to interpret nitrogen balance and soil water reserve under
different tillage managements. Journal of Water and Land Development, 2014, 22 (1), pp.33-39.
<10.2478/jwld-2014-0020>. <hal-01143023>
HAL Id: hal-01143023
https://hal.archives-ouvertes.fr/hal-01143023
Submitted on 20 Apr 2015
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entific research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destin´ee au d´epˆot et `a la diffusion de documents
scientifiques de niveau recherche, publi´es ou non,
´emanant des ´etablissements d’enseignement et de
recherche fran¸cais ou ´etrangers, des laboratoires
publics ou priv´es.
© Polish Academy of Sciences (PAN) in Warsaw, 2014; © Institute of Technology and Life Science (ITP) in Falenty, 2014
© Polish Academy of Sciences, Committee for Land Reclamation JOURNAL OF WATER AND LAND DEVELOPMENT
and Environmental Engineering in Agriculture, 2014 J. Water Land Dev. 2014, No. 22 (VII–IX): 33–39
© Institute of Technology and Life Science, 2014 PL ISSN 1429–7426
Available (PDF): www.itp.edu.pl/wydawnictwo; http://www.degruyter.com/view/j/jwld
Received 22.05.2014
Reviewed 11.07.2014
Accepted 14.07.2014
A – study design
B – data collection
C – statistical analysis
D – data interpretation
E – manuscript preparation
F – literature search
Using Water and Agrochemicals in the soil, crop
and Vadose Environment (WAVE) model
to interpret nitrogen balance and soil water reserve
under different tillage managements
Narjes ZARE
1) BCDEF
, Mohammadreza KHALEDIAN
1), 2) ABCDEF
Jean-Claude MAILHOL
2) ABD
1)
University of Guilan, Faculty of Agricultural Sciences, Irrigation and Drainage Department, Rasht, Iran;
e-mail: khaledian@guilan.ac.ir
2)
National Research Institute of Science and Technology for Environment and Agriculture, Irstea, Montpellier, France;
e-mail: mkh572000@yahoo.com
For citation: Zare N., Khaledian M., Mailhol J.C. 2014. Using Water and Agrochemicals in the soil, crop and Vadose En-
vironment (WAVE) model to interpret nitrogen balance and soil water reserve under different tillage man-
agements. Journal of Water and Land Development. No. 22 p. 33–39.
Abstract
Applying models to interpret soil, water and plant relationships under different conditions enable us to
study different management scenarios and then to determine the optimum option. The aim of this study was us-
ing Water and Agrochemicals in the soil, crop and Vadose Environment (WAVE) model to predict water con-
tent, nitrogen balance and its components over a corn crop season under both conventional tillage (CT) and di-
rect seeding into mulch (DSM). In this study a corn crop was cultivated at the Irstea experimental station in
Montpellier, France under both CT and DSM. Model input data were weather data, nitrogen content in both the
soil and mulch at the beginning of the season, the amounts and the dates of irrigation and nitrogen application.
The results show an appropriate agreement between measured and model simulations (nRMSE < 10%). Using
model outputs, nitrogen balance and its components were compared with measured data in both systems. The
amount of N leaching in validation period were 10 and 8 kg·ha
–1
in CT and DSM plots, respectively; therefore,
these results showed better performance of DSM in comparison with CT. Simulated nitrogen leaching from CT
and DSM can help us to assess groundwater pollution risk caused by these two systems.
Key words: conservation tillage, corn, modeling, nitrogen balance, soil water reserve
INTRODUCTION
Irrigation plays a dominant role in agriculture be-
cause of degrading water resources and variant distri-
bution of the rainfalls in Mediterranean region
throughout the year. Farmers need to save water and
make judicious use of it, especially during the dry
season. So, improving irrigation management is cru-
cial to address water scarcity issue [A
DEKALUE, FAPO-
HUNDA 2006]. Nowadays, understanding the proc-
esses which affect the fate of nitrogen in the root zone
is essential to develop relevant management strategies
to conserve surface and subsurface water resources.
The application of mechanistic nitrogen models to
describe the nitrogen fluxes in the soil–plant–
atmosphere continuum helps us to better understand
the soil nitrogen balance.
In the traditional soil management, burning of
crop residues and declining of fallow periods in large-
scale will reduce water infiltration and nutrient imbal-
ances [O
UEDRAOGO et al. 2004]. Some activities like
mechanized seedbed preparation, overgrazing by cat-
tle little would reduce abundance and biodiversity of
soil organisms [B
ROWN et al. 2001]. One of the envi-
DOI: 10.2478/jwld-2014-0020
Brought to you by | IRSTEA
Authenticated
Download Date | 2/26/15 2:27 PM
34 N. ZARE, M. KHALEDIAN, J.C. MAILHOL
© PAN in Warsaw, 2014; © ITP in Falenty, 2014; J. Water Land Dev. No. 22 (VII–IX)
ronmental conservation system is DSM (Direct Seed-
ing into Mulch).This approach made conserve the
soil, increasing crop yield and improve the environ-
mental conditions and soil ecology [E
RENSTEIN
2003]. DSM also saves the soil water by decreasing
evaporation and runoff. By providing favorable food
source, the activity of soil micro-organisms increases
which results in degradation of organic matter, and
ends up in the mineralization of nitrogen, N
[S
CHROTH et al. 2001].
To evaluate different aspects of DSM, field ex-
periment can help us but field trials are costly and
time-consuming, therefore using of crop models is
recommended to save time and money. Also it made
it possible to use other advantages such as examining
of different management scenarios. Models are suit-
able tools to reestablish the real systems under physi-
cal, chemical and biological conditions; however, in-
put supporting faced some difficulties [A
DDISCOTT
2000]. Different models were developed to better un-
derstand the water and solute transportation in soil
[S
IMUNEK et al. 1999]. Crop models can predict crop
growth and their development under different climate,
agricultural practices, soil features and environmental
conditions. Models are efficient tools for water man-
agement [P
ANDA et al. 2003]. Many models were in-
troduced by researchers among them some models are
multi-crop and others are single crop e.g., ORYZA
[P
ANDA et al. 2003], WARM [CONFALONIERI et al.
2010], CERES-Wheat [OTTER-NACKE et al. 1987].
Field-tested computer models can be useful tools
of assessing nutrient leaching and water movement. In
particular, when data are scarce or of limited reliabil-
ity, physically-based numerical models can be useful
exploratory tools to understand the complexity of
these processes and to quantify nitrogen leaching as a
consequence of fertilizer practices. One of the nu-
merical models in this field is WAVE (Water and Ag-
rochemicals in the soil and Vadose Environment)
model, developed and introduced by V
ANCLOOSTER et
al. [1994]. This model was evaluated to predict crop
water consumption and volumetric soil water content
in a cropped soil during 1992 and 1993. The results
showed that the model is reliable to simulate the fate
of water in a sandy soil and in semi-arid conditions
[F
ERNANDEZ et al. 2002]. In Bulgaria, water flow and
nitrogen transport also were simulated by other mod-
els i.e. SWAP and ANIMO, to optimize agricultural
practice aiming to minimize the impact on the envi-
ronment. The results showed that SWAP model could
satisfactorily simulated soil water dynamics, while
simulations of ANIMO for nitrogen cycle present
greater divergence with observations, but in assessing
land use impact on groundwater quality had adequate
precision [M
ARINOV et al. 2005].
The purpose of this study was to assess the per-
formance of WAVE model in describing soil water
and soil nitrogen balances on a loamy soil under both
conventional tillage and direct seeding into mulch
systems. This topic was identified as being important
to introduce a relevant tool to better understand DSM
system.
MATERIAL AND METHODS
EXPERIMENTAL SITE
An experimental study has been carried out to
evaluate the N and water balances of corn (Pioneer
PR35Y65 variety) in 2007 with DSM and CT. This
study was conducted at Irstea experimental station
which is located in Montpellier in the southeastern
part of France in a Mediterranean climate with 789
mm of average annual rainfall (1991–2007). The an-
nual evapotranspiration (ET
o
) over this long term cal-
culated by Penman–Monteith equation was 859
mm·year
–1
. The climate data were monitored at the
weather station situated in the experimental station.
Figure 1 shows average monthly precipitation and ET
o
(mm) at the Irstea institute (1991–2007). Figure 1
demonstrating that in corn crop season in the region
(April–September), precipitation is lower than ET
o
so
that irrigation is necessary. According to the USDA
soil classification, the soil under CT and DSM plots
belongs to the loam soil category. The values of
chemical properties of the soil are given in Table 1.
0
30
60
90
120
150
180
01234567891011121314
Precipitation and ETo , mm
Months of year
ETo
Precipitation
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Fig. 1. Average monthly precipitation and reference
evapotranspiration (ET
o
, mm) at the Irstea institute
(1991–2007); source: own study
Table 1. Soil chemical properties at the Lavalette Agri-
cultural Research Station in both conventional tillage (CT)
and direct seeding into mulch (DSM) plots
Organic matter Organic carbon N total
Plot
%
C:N
CT 1.39 0.78 0.08 10
DSM 1.79 1.04 0.09 11.81
Presented soil properties are for 0–30 cm depth.
Source: own study.
EXPERIMENTAL SETUP
Two main plots were considered in this study:
a DSM plot of about one hectare and a CT plot of 1.7
ha. The corn planting and the harvest dates were 24
April 2007 and 28 September 2007, respectively. For
a given crop season, the crop is generally subjected to
different irrigation treatments, with always at least
a full irrigated (430 mm of irrigation depth) and
Brought to you by | IRSTEA
Authenticated
Download Date | 2/26/15 2:27 PM
Using Water and Agrochemicals in the soil, crop and Vadose Environment (WAVE) model to interpret nitrogen balance… 35
© PAN in Warsaw, 2014; © ITP in Falenty, 2014; J. Water Land Dev. No. 22 (VII–IX)
a rain-fed treatments. Other treatments were respec-
tively 218 and 182 mm of irrigation depth. An access
tube of neutron probe was installed in each treatment.
Soil water reserve (SWR) was monitored once a week
using a neutron probe from 0 to 2 m at 0.1 m depth
interval.
A corn crop was cultivated in CT and DSM. For
maintaining a thick layer of mulch on the soil surface
in DSM, a mixed of oat, vetch and rape seed was
sown in October 2006 in the DSM as cover crop and
was destroyed by “glyphosate (Rounup)” in April
2007, before corn sowing (two weeks before sowing).
The amount of residues from precedent crops at sow-
ing and produced mulch by cover crop were 0.5 and
8 Mg·ha
–1
, respectively. The field was fallow in the
middle season of 2007 and 2008 with CT. The model
was validated in this period to simulate soil water
content and nitrogen balance components.
The agricultural practices and the use of plant
protection agents were in accordance with local prac-
tices, official recommendations and experts’ advice.
The goal was to mimic as close as possible the condi-
tions of production in commercial farms. So that, the
farm scale equipment was fixed and applied. In the
CT plot, at the end of July disc harrow was used to
chop and bury the residues of the precedent crops. In
the middle of November, tillage with plough was per-
formed. In DSM plot, the cover crop was sown by
a specific seeder. After destroying the cover crop the
same seeder was used to sow the main crop. N
amounts were applied in order to fully satisfy plant
requirement as soon as an N soil profile was estab-
lished just before sowing. Two N applications were
generally performed, the first one at sowing and the
second one 30–40 days after sowing (DAS). Other
fertilizers (P and K) also were applied in both plots.
To determine the grain yield (GY) and dry matter
yield (DM) ten 3 m
2
sub-plots were hand harvested
after reaching physiological maturity. The measured
GY and DM coefficient of variation (coefficient of
variation, CV) ranged from 6 to 12%.
The soil mineral N content was determined be-
fore sowing and after harvest. Seven core samples per
plot were taken from 0 to 0.10, 0.10 to 0.30, 0.30 to
0.60, 0.60 to 0.90, 0.90 to 1.20, and 1.20 to 1.50 m
depths with an auger. Then, the samples of each layer
were mixed and sieved to have a representative and
a unique sample for each layer. Plant N content in
leaves, stems and grains were determined at harvest
with the Kjeldahl method.
THE NUMERICAL MODEL
The WAVE model [VANCLOOSTER et al. 1994] is
a deterministic and mathematical model that simulates
the transfer and the fate of agrochemicals and the
movement of water in the soil-crop continuum. This
model combines two models: the SWATNIT model
[V
EREECKEN et al. 1991] and SUCROS model [SPIT-
TERS
et al. 1988]. Because of inactivation of SU-
CROS model in current WAVE model version, in this
study PILOTE model [K
HALEDIAN et al. 2009] was
used instead of SUCROS model. This model simu-
lates the soil water balance and the crop yield at
a daily time step under the assumption of just water as
growth limiting factor [M
AILHOL et al. 1997]. This
model was calibrated and validated in the same field
for corn in both CT and DSM systems by K
HALEDIAN
et al. [2009].
The WAVE model uses finite difference tech-
niques for the solution of the physical transport equa-
tions [T
IMMERMAN, FEYEN 2003].The modules cur-
rently are available in the WAVE model simulate
with the following soil processes: the flow of water,
transport of non-reactive solutes, the heat transport,
the crop growth and the movement and transforma-
tions of nitrogen. It can deal with different soil hori-
zons which are divided into equidistant soil compart-
ments. A water, heat and solute mass balance equa-
tions are developed for each section, taking into ac-
count different sink/source terms.
The water transport module of WAVE is based
on the Richards equation for isotropic, homogeneous,
isothermal, rigid and porous media, while solute
transport was modeled with a convection equation:
(1)
where:
C(h)
K(h)
h
Sinkwat
z
t
– the differential moisture capacity;
– the hydraulic conductivity relationship;
– the soil water pressure head, m;
– the water sink term, 1·s
–1
;
– the space co-ordinates, m;
– the time co-ordinates, s.
The water transport model is assumed that due to
it the soil water flow occurs in response to a hydraulic
potential gradient. The soil water retention curve is
assumed to be of the form given by Van G
ENUCHTEN
[1980]:
(2)
where:
α
m, n
– the saturated moisture content;
– the inverse of entry value, 1·m
–1
;
– shape parameters.
Solute transport was described by the convection-
dispersion equation with a linear reversible adsorption
isotherm for reactive solute:
(3)
where:
C
r
ρ
– the resident concentration (kg·m
–3
) in the
solution;
– the soil bulk density, kg·m
–3
;
Brought to you by | IRSTEA
Authenticated
Download Date | 2/26/15 2:27 PM
36 N. ZARE, M. KHALEDIAN, J.C. MAILHOL
© PAN in Warsaw, 2014; © ITP in Falenty, 2014; J. Water Land Dev. No. 22 (VII–IX)
K
d
D
s
q
Σ
i
φ
i
– the solute distribution coefficient, m
3
·kg
–1
;
– the apparent dispersion coefficient, m
2
·s
–1
;
– the Darcian water flux, m·s
–1
;
– a solute sink term (kg·m
–3
·s
–1
), including
crop uptake and transformation.
When solute is applied at the soil surface (during
a fertilization or irrigation event), it is assumed to dis-
solve instantaneously in the mass of water entering
the soil profile during the day of solute application (or
the first day when infiltration occurs).
WAVE simulates plant uptake under tension
condition by determining the fraction of the total
growing season of potential uptake and the calculated
convective plant water uptake. Mineralization of or-
ganic N and immobilization of inorganic N were
modeled based on soil litter, soil humus and soil ma-
nure pools. In addition, potential rates for modeling
such as nitrogen transformations were reduced for the
temperature and the water content in the soil profile.
NITROGEN BALANCE
Modeling flow and nitrogen transport with
WAVE model need multiple parameters to be esti-
mated; however, collecting all parameters and direct
estimation really are impossible. Numerical models
are often used instead of inverse modeling and ex-
perimental method to solve multiple parameter prob-
lems [R
ITTER et al. 2004].
For experimental measurements, a simple method
to assess the N balance in both systems was used. N
mineralization was estimated in zero N application
plots [S
EXTON et al. 1996]:
N
Min
= N
F
– N
I
+ N
P
+ N
MF
– N
MI
(4)
where N
I
and N
F
are the soil N contents in the begin-
ning and in the end of the study period, respectively.
N
P
is the N uptake by plant. N
MI
and N
MF
are mulch N
content at the beginning and at the end of study pe-
riod. In CT, N
MF
and N
MI
were not considered. A loss
of N in the fertilized plots (P) was calculated with
equation 5 using N
Min
, N
I
, N
F
, N
P
, N
MF
, and N
MI
as
defined above and N
AP
(N fertilizer applied). A nega-
tive value of P is interpreted as a loss of N in the soil–
plant system.
P = N
F
– N
I
+ N
P
– N
AP
– N
Min
+ N
MF
– N
MI
(5)
MODEL CALIBRATION
Sensitivity analyses are valuable tools for identi-
fying important model parameters, testing the model
conceptualization, and improving the model structure.
They help to apply the model efficiently and to enable
a focused planning for future research and field meas-
urement. Sensitivity analysis of WAVE model by
M
UNOZ-CARPENA et al. [2008] showed that leaf area
index, distribution coefficient of NH
4
+
and maximum
N uptake had a medium sensitivity level, where crop
coefficient, saturated soil water content and curve
shape parameter had a high sensitivity level.
For illustrating the impact of using effective cali-
brated field parameters instead of laboratory scale
parameters, calibrated modeling results are shown as
well. Calibration was performed on a trial and error
basis, using the field scale observed moisture content
and nitrogen balance as objects (Tab. 2).
Table 2. Summary of simulation results (kg·ha
–1
) in conventional tillage (CT) and direct seeding into mulch (DSM) systems
CT DSM
calibration validation calibration validation
Output
obs. sim. obs. sim. obs. sim. obs. sim.
Final NO
3
–
148.7 149.0 140.9 140.5 328.2 327.4 67 70.8
Final NH
4
+
29.5 29.8 25.5 34 24.5 25.2 27 20.2
N leaching 16 17 7 10 6 4 7 8
N mineralization 110 99 110 109 305 299 160 150
Final stock simulated in 150 cm soil.
Explanations: obs. – observation, sim. – simulation.
Source: own study.
To express the differences between the simulated
and observed values in terms of statistical indices, the
normalized root mean square error, nRMSE, MAE
being the mean absolute error, as well as, the coeffi-
cient of residual mass, CRM were used as follows:
(6)
(7)
(8)
where:
P
i
, O
i
n
– the model calculated and observed val-
ues, respectively;
– the number of samples;
– the mean of the observed values.
These statistical indices were calculated on un-
sorted data, observed and predicted values being
compared directly. nRMSE, MAE and CRM should
be as close as possible to zero. The simulation is con-
sidered excellent with nRMSE less than 10%, good if
nRMSE is greater than 10 and less than 20%, fair if
nRMSE is greater than 20 and less than 30%, and
Brought to you by | IRSTEA
Authenticated
Download Date | 2/26/15 2:27 PM
Using Water and Agrochemicals in the soil, crop and Vadose Environment (WAVE) model to interpret nitrogen balance… 37
© PAN in Warsaw, 2014; © ITP in Falenty, 2014; J. Water Land Dev. No. 22 (VII–IX)
poor if nRMSE is greater than 30% [BANNAYAN,
HOOGENBOOM 2009]. The coefficient of residual
mass, CRM, ranges between –infinite and + infinite,
with the optimum = 0. If it is positive, it indicates that
the model underestimates the prediction, if negative
indicates overestimation and when it is close to zero
indicates the absence of trends [B
ONFANTE et al.
2010].
RESULTS AND DISCUSSION
After running the model, N balance parameters
were simulated (Tab. 2). In this table, observation and
simulation values in two plots were compared. Final
N content that consists of NO
3
–
and NH
4
+
, had suit-
able simulation because of closed value of simulation
to observation. The most important parameter in this
table is N leaching. By comparing this parameter in
CT and DSM plots, derive less value in DSM plot. In
some references underlined that in DSM plot, runoff
and leaching have enhanced because the crop roots
have disintegrated after cutting the crop [K
HALEDIAN
2009], so macro pore from decomposed root make
increase the deep percolation ended up in N leaching
increase. But the results showed another thing. With
a comparison of corresponding N leaching in two
treatments, it is understood that DSM had better ap-
plication of saving water and nitrogen, because of
lower deep percolation and N leaching. The cause of
this subject should be understood in soil chemical
properties. In Table 1, values of organic carbon and
organic matter in DSM and CT plots showed that
DSM has higher value while organic carbon in CT
and DSM were 0.78 and 1.04% and organic matter
were 1.39 and 1.79%, respectively. Organic matter
has involved in particle connections and reduction of
soil macro pore so, the soil become more stable. Ac-
cording to this process, DSM has higher water saving
capacity and lower N leaching. On the other hand, the
root of cover crop absorbs residual nitrogen of soil
and composed mulch will be released N being suit-
able for main crop. Therefore, DSM has an important
role to save soil water and nitrogen.
In similar studies, WAVE model was used to
simulate the fate of nitrogen following in France. In
calibration process, modelled budget error was arisen
from net mineralization of +15 and +9 kg N·ha
–1
for
controlled plot and cropping season on a plot, respec-
tively. These values in validation were –9 and +13 kg
N·ha
–1
which was considered as very acceptable
[PAYET et al. 2009]. DUWING et al. [2003] used this
model to evaluate its prediction capabilities. So water
and NO
3
–
concentrations and fluxes were monitored
in two sites (New Caledonia and France) with differ-
ent surface covers (maize or bare soil) in three con-
secutive years. For both sites, evaluations of the mod-
el showed the best results for wet conditions.
In addition to nitrogen, SWR was simulated by
WAVE. In Figures 2 and 3, observed and simulated
values were compared in both plots. Figure 2 shows
0
100
200
300
400
500
100 150 200 250 300
SWR , mm
Day number of the year
WAVE Measurement
0
100
200
300
400
28/3 6/ 7 14/10 22/1 2/ 5 10/8 18/11
SWR , mm
Date
WAVE Measurement
28/3 6/7 14/10 22/1 2/5 10/8 18/11
Fig. 2. Comparison between simulation and measurement values of soil water reserve (SWR) in direct seeding into mulch
(DSM) plot. (Calibration (left) and Validation (right)); source: own study
80
130
180
230
280
330
380
100 150 200 250 300
SWR , mm
Day number of the year
WAVE Measurement
Fig. 3. Comparison between simulation and measurement of
soil water reserve (SWR) in conventional tillage (CT) plot;
source: own study
this comparison in calibration and validation proc-
esses in DSM plot. Left-hand side results from com-
paring of calibrated and observed values present a
satisfied simulation of this parameter. Also validated
data were so close to SWR measurements (right-hand
side). Also in CT plot calibration was done. Figure 3
shows the good correlated between the two data se-
ries. Because of improper condition in bare field to
measure SWR with neutron probe, measurements did
not perform in CT plot. So, model evaluation in CT
plot consists of calibration without validation.
Statistical indices should be used to analyze the
simulated data. In Table 3 model assessment was
done by CRM, nRMSE and MAE. In calibration both
Brought to you by | IRSTEA
Authenticated
Download Date | 2/26/15 2:27 PM
38 N. ZARE, M. KHALEDIAN, J.C. MAILHOL
© PAN in Warsaw, 2014; © ITP in Falenty, 2014; J. Water Land Dev. No. 22 (VII–IX)
plots have low value of CRM. nRMSEs in all cases
were less than 10%. According to acceptable values
of each one, WAVE had an excellent simulation in
CT and DSM plots according to nRMSE classification
[B
ANNAYAN, HOOGENBOOM 2009]. Also assessment
of the MAE values (6.3% for CT calibration and 4.6%
for DSM calibration) illustrates suitable simulation of
the model.
Table 3. Model soil water reserve (mm) simulation assess-
ment in conventional tillage (CT) and direct seeding into
mulch (DSM) plots
Item Plot
RMSE
mm
nRMSE
%
MAE
%
CRM
CT 15 7 6.3 0.022
Calibration
DSM 15 5 4.6 0.017
Validation DSM 9 3 7.9 0.023
Source: own study.
CONCLUSIONS
Water and nitrogen are important inputs in agri-
cultural production systems, especially in Mediterra-
nean climate because of erratic rainfalls. So, selecting
the most appropriate cropping system, irrigation and
nitrogen application managements to save those items
is crucial. In this study, calibration and validation of
the WAVE model in CT and DSM plots were done by
analyzing of N balance and SWR. The results show
that saving soil water and nitrogen in DSM plot is
more effective than CT plot. In CT and DSM plot,
simulated values of SWR were compared to meas-
urements. That comparison showed a good agreement
between simulated and measurement values according
to statistical indices. CRM values were close to zero.
Depending on low nRMSE values (less than 10%),
the simulations were excellent in CT and DSM plots.
According to the obtained results, Wave model can be
used as a relevant tool to manage water and nitrogen
in both CT and DSM systems to save water and nitro-
gen and mitigate negative environment effects.
REFERENCES
ADEKALU K.O., FAPOHUNDA H.O. 2006. A numerical model
to predict crop yield from soil – water deficit. Biosystems
Engineering. Vol. 94(3) p. 359–372.
A
DDISCOTT T.M. 2000. Tillage, mineralization and leaching.
Soil and Tillage Research. Vol. 53 (3–5) p. 163–165.
BANNAYAN M., HOOGENBOOM G. 2009. Using pattern recog-
nition for estimating cultivar coefficients of a crop simu-
lation model. Field Crops Research. Vol. 111 p. 290–302.
B
ONFANTE A., BASILE A., ACUTIS M., DE MASCELLIS R.,
MANNA P., PEREGO A., TERRIBILE F. 2010. Swap, cropsyst
and macro comparison in two contrasting soils cropped
with maize in northern Italy. Agricultural Water Man-
agement. Vol. 97 p. 1051–1062.
BROWN G., PASINI A., BENITO N.P., AQUINO A.M., CORREIA
E. 2001. Diversity and functional role of soil macrofauna
communities in Brazilian no tillage agroecosytems. In:
International symposium on managing biodiversity in ag-
ricultural ecosystems. Ed. R. Lal. 8–10 November 2001,
Montreal, Canada. IISD p. 19.
C
ONFALONIERI R., BELLOCCHI G., TARANTOLA S., ACUTIS M.,
DONATELLI M., GENOVESE G. 2010. Sensitivity analysis
of the rice model warm in Europe: exploring the effects
of different locations, climates and methods of analysis
on model sensitivity to crop parameters. Environmental
Modelling and Software. Vol. 25 p. 479–488.
D
UWING C., NORMAND B., VAUCLIN M., VACHAUD G., GREEN
S., BECQUER T. 2003. Evaluation of the WAVE model for
predicting nitrate leaching for two contrasted soil and
climate conditions. Vadose Zone Journal. Vol. 2(1) p. 76–
89.
ERENSTEIN O. 2003. Smallholder conservation farming in the
tropics and subtropics: a guide to the development and
dissemination of mulching with crop residues and cover
crops. Agriculture, Ecosystems and Environment. Vol.
100 p. 17–37.
F
ERNANDEZ J.E., SLAWINSKI C., MORENO F., WALCZAK R.T.,
VANCLOOSTER M. 2002. Simulating the fate of water in
a soil–crop system of a semi-arid Mediterranean area with
the WAVE 2.1 and the EURO-ACCESS-II models.
Agricultural Water Management. Vol. 56 p. 113–129.
KHALEDIAN M.R. 2009. Evaluation de la technique du semis
direct en culture irriguee en comparaison avec le systeme
de culture conventionnel. Dissertation. University of
Montpellier II pp. 138.
KHALEDIAN M.R., MAILHOL J.C., RUELLE P., ROSIQUE P.
2009. Adapting pilote model for water and yield man-
agement under direct seeding system: the case of corn and
durum wheat in a Mediterranean context. Agricultural
Water Management. Vol. 96 p. 757–770.
M
AILHOL J.C., OLUFAYO A.A., RUELLE P. 1997. Sorghum and
sunflower evapotranspiration and yield from simulated
leaf area index. Agricultural Water Management. Vol. 35
p. 167–182.
MARINOV D., QUERNER E., ROELSMA J. 2005. Simulation of
water flow and nitrogen transport for a Bulgarian experi-
mental plot using SWAP and ANIMO models. Journal of
Contaminant Hydrology. Vol. 77 p. 145–164.
MUNOZ-CARPENA R., RITTER A., BOSCH D.D., SCHAFFER B.,
POTTER T.L. 2008. Summer cover crop impacts on soil
percolation and nitrogen leaching from a winter corn
field. Agricultural Water Management. Vol. 95 p. 633–
644.
OTTER-NACKE S., GODWIN D.C., RITCHIE J.T. 1987. Testing
and validating the ceres-wheat model in the diverse envi-
ronments. Agristars publ. No. Ym-15-00407. Ntis,
Springfield, va. pp. 147.
O
UEDRAOGO E., MANDO A., BRUSSAARD L. 2004. Soil macro-
faunal-mediated organic resource disappearance in semi-
arid West Africa. Applied Soil Ecology. Vol. 27 p. 259–
267.
P
ANDA R.K., BEHERA S.K., KASHYAP P.S. 2003. Effective
management of irrigation water for wheat under stressed
conditions. Agricultural Water Management. Vol. 63(1)
p. 37–56.
PAYET N., FINDELING A., CHOPART J-L., FEDER F., NICOLINI
E., SAINT MACARY H., VAUCLIN M. 2009. Modelling the
fate of nitrogen following pig slurry application on a trop-
ical cropped acid soil on the island of reunion (France).
Agriculture, Ecosystems and Environment. Vol. 134 p.
218–233.
R
ITTER A., MUNOZ-CARPENA R., REGALADO C.M., VAN-
CLOOSTER M., LAMBOT S. 2004. Analysis of alternative
measurement strategies for the inverse optimization of the
Brought to you by | IRSTEA
Authenticated
Download Date | 2/26/15 2:27 PM
Using Water and Agrochemicals in the soil, crop and Vadose Environment (WAVE) model to interpret nitrogen balance… 39
© PAN in Warsaw, 2014; © ITP in Falenty, 2014; J. Water Land Dev. No. 22 (VII–IX)
hydraulic properties of a volcanic soil. Journal of Hydrol-
ogy. Vol. 295 p. 124–139.
SCHROTH G., SALAZAR E., DA SILVA Jr. J.P. 2001. Soil nitro-
gen mineralization under tree crops and a legume cover
crop in multi-strata agroforestry in central Amazonia:
spatial and temporal patterns. Experimental Agriculture.
Vol. 37 p. 253–267.
S
EXTON B.T., MONCRIEF J.F., ROSEN C.J., GUPTA S.C., CHENG
H.H. 1996. Optimizing nitrogen and irrigation inputs for
corn on nitrate leaching and yield on a Coarse-Textured
soil. Journal of Environmental Quality. Vol. 25 p. 982–
992.
S
IMUNEK J., SEJNA M., VAN GENUCHTEN M.Th. 1999. The
hydrus-2d software package or simulating the one-
dimensional movement of water, heat and multiple sol-
utes in variably-saturated media. Ver. 2.0. IGWMC-TPS
53. International Ground Water Modeling Center, Colo-
rado School of Mines, Golden, Colorado pp. 251.
SPITTERS C.J.T., VAN KEULEN H., VAN KRAAILINGEN D.W.G.
1988. A simple but universal crop growth simulation
model, surcros87. In: Simulation and systems manage-
ment in crop protection. Ed. R. Rabbinge, S.A. Ward,
H.H. van Laar. Wageningen, The Netherlands. Pudoc.
Simulation monographs. Vol. 32 p. 147–181.
T
IMMERMAN A., FEYEN J. 2003. The wave model and its ap-
plication; simulation of the substances water and agro-
chemicals in the soil, crop and vadose environment. Re-
vista Corpoica. Vol. 4(1) p. 36–41.
VAN GENUCHTEN M.Th. 1980. A closed-form equation for
predicting the hydraulic conductivity of unsaturated soils.
Soil Science Society America Journal. Vol. 44 p. 892–
898.
V
ANCLOOSTER M., VIAENA J., CHRISTIAENS K. 1994. Wave
a mathematical model for simulating water and agro-
chemicals in the soil and vadose environment. Reference
and user’s manual, release 2.0.
VEREECKEN H., VANCLOOSTER M., SWERTS M., DIELS J. 1991.
Simulating water and nitrogen behaviour in soil cropped
with winter wheat. Fertilizer Research. Vol. 27 p. 233–
243.
Narjes ZARE, Mohammadreza KHALEDIAN, Jean-Claude MAILHOL
Użycie modelu Water and Agrochemicals in the soil, crop and Vadose Environment (WAVE)
do interpretacji bilansu azotu i glebowych zasobów wody w różnych warunkach uprawy
STRESZCZENIE
Słowa kluczowe: bilans azotu, konserwująca uprawa gleby, modelowanie, zasoby wody w glebie, zboża
Zastosowanie modeli do interpretacji zależności, które zachodzą w różnych systemach uprawy między gle-
bą, wodą i roślinami umożliwia zbadanie odmiennych scenariuszy gospodarowania, a następnie wybór optymal-
nej opcji. Celem badań było zastosowanie modelu WAVE (Water and Agrochemicals in the soil, crop and Vado-
se Environment) do prognozowania zawartości wody, bilansu azotu i jego form w sezonie wegetacyjnym w wa-
runkach konwencjonalnej uprawy (CT) i siewu bezpośrednio w mulcz (DSM). Oba systemy zastosowano do
uprawy zbóż w stacji doświadczalnej Irstea w Montpellier we Francji. Danymi wejściowymi do modelu były
warunki pogodowe, zawartość azotu w glebie i w mulczu na początku sezonu wegetacyjnego, dawki i terminy
nawodnień oraz nawożenie azotem. Wyniki potwierdzają zgodność pomiarów i symulacji modelowych (nRMSE
< 10%). Porównano bilans azotu i jego składników uzyskane za pomocą modelu i na podstawie danych z bezpo-
średnich pomiarów w obu wariantach upraw. Ilość wymywanego azotu w okresie badań wynosiła 10 (system
CT) i 8 kg·ha
–1
(system DSM). Taki wynik dowodzi korzystniejszego oddziaływania systemu DSM w porówna-
niu z systemem CT. Symulacja wymywania azotu z upraw w systemie CT i DSM umożliwia ocenę ryzyka za-
nieczyszczenia azotem wód gruntowych w wyniku stosowania obu systemów uprawy.
Brought to you by | IRSTEA
Authenticated
Download Date | 2/26/15 2:27 PM
