Computers and Electronics in Agriculture 121 (2016) 301-312

Contents lists available at ScienceDirect

Computers and Electronics in Agriculture

journal homepage : www .elsevier .com /locate /compag

Original papers

The combination of geostatistics and geochemical simulation for the

site-specific management of soil salinity and sodicity

Faranak Ranjbar , Mohsen Jalali

Department of Soil Science, Faculty of Agriculture, Bu-Ali Sina University, Hamedan, Iran

abstract
article

info
The spatial variability of soil parameters such as salinity and sodicity is very effective in determining the

Article history:
Received 23 June 2015
Received in revised form 17 November 2015
Accepted 20 December 2015

Keywords:
Spatial variability
Geochemical simulation
Sodic water
PHREEQC

suitable areal cropping pattern and appropriate management of agricultural lands. To conduct this study,
a grid sampling at 100 m was carried out in an agricultural area of 30 ha located in Tajarak of Hamedan,
western Iran. After physicochemical analyses of soil samples, soil properties which had significant correlations with the exchangeable sodium percentage (ESP) were identified using the statistical analysis. The
ESP was significantly positively correlated with the clay content, pH, electrical conductivity (EC), cation
exchange capacity (CEC), and sodium adsorption ratio (SAR), while it showed a significant negative correlation with the silt content. The spatial variability of ESP and its related parameters investigated by the
geostatistical analysis showed that pH, EC, CEC, SAR, and ESP were strongly spatially dependent, while
the clay and silt contents had a moderate spatial dependency. The distribution maps indicated that most
soils in the study area did not have a sodicity problem, while ESP of soils in some parts was more than
15%. The long-term impact of the application of saline-sodic and non-saline-sodic irrigation waters on
some chemical properties of a saline sodic soil under 20 years of the continuous cultivation of wheat
and alternate cultivation of wheat and sugar beet was simulated using the geochemical PHREEQC program. The results showed that the chemical quality of the soil was significantly improved after 20 years
of irrigation with non-saline-sodic water. On the other hand, the problem of the soil salinity was accelerated after 20 years of irrigation with saline-sodic water, while ESP remained unchanged. Due to the
high water requirement of sugar beet, the effect of its cultivation was not significant in reducing soil sodicity despite the removal of Na + from the soil by this crop. Therefore, it is recommended that the cropping
pattern should be changed by planting crops with low water requirements or rainfed crops if sodic irrigation water is applied.
2015 Elsevier B.V. All rights reserved.

1. Introduction
Salinity and sodicity are two of the most destructive processes in
soils, particularly in arid and semi-arid areas. The investigation of
changes in the soil salinity and sodicity, especially in large areas, is
expensive and time-consuming. Hence, it is necessary to use
techniques that can easily monitor a wide range of areas and
examine the changes in the soil salinity and sodicity.
One of the key characteristics of soils is the spatial and temporal
variability. There are several factors involved in the process of the
soil formation that vary over the time and influence soil properties.
On the other hand, the management of agricultural lands is impossible based on a set of discrete sampled points as these data must
be converted to a continuous format of information. Therefore, it is
necessary to apply the mathematical and statistical data analyzing
Corresponding author.
E-mail address: f_ranjbar1980@yahoo.com (F. Ranjbar).
http://dx.doi.org/10.1016/j.compag.2015.12.010
0168-1699/ 2015 Elsevier B.V. All rights reserved.

methods able simultaneously to use the quantitative and

numerical information of variables and information related to their
geographical location.
The soil variability in the field is usually described by classical
statistical methods and is assumed to be random (Cemek et al.,
2007). The soil variability results from effects and interactions of
different processes in the soil profile (Parkin, 1993). Webster
(1985) indicated that soil properties are usually spatially dependent. Samples close to each other have more similar properties than
those further away from each other. The spatial variability of surface soil properties (Brejda et al., 2000), amount of soil nutrients
(Newman et al., 1997), soil chemical conditions (Lee et al., 2001),
nitrate leaching (Ersahin, 1999), distribution of pesticides in the soil
(Rao and Wagenet, 1985), and release of potassium (K+) from some
calcareous soils (Jalali, 2007a) has been analyzed using geostatistical methods. Information on the spatial variability of soil properties
is necessary to implement a site-specific soil management, e.g., a
variable rate of the application of fertilizers (Wei et al., 2009).

302

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

The geostatistical technique is used to demonstrate the spatial

dependency of soil properties, both isotropically and anisotropically (Burgess and Webster, 1980). Furthermore, the proper geostatistical analysis provides valuable information about the spatial
distribution of soil properties in agricultural lands (Liu and Yang,
2008). Geostatistics can provide a description and modeling of spatial patterns, prediction at unsampled locations, and assessment of
uncertainty of these predictions by using a set of statistical tools.
Studies related to the spatial variability of soil properties have
been performed since the 1970s. Geostatistics, based on the regional variables, has been proven to be one of the most effective
methods to analyze the spatial distribution and variation of soil
properties (Liu and Yang, 2008). A wide range of sampling points
with different distances has been used in different studies, e.g.,
33 sampling points in a representative of 929 ha (Shifteh Somee
et al., 2011), 40 sampling points from an agricultural field of
30 ha (Jalali, 2007a), 60 sampling points in a representative of
8.187 ha (Cemek et al., 2007), and 80 and 84 sampling points from
areas of 11.2 and 6 ha, respectively (Lopez-Granados et al., 2002).
Sustainable agriculture is a potentially useful tool in meeting
the future needs of a growing world population for food. The production of agricultural crops is difficult in saline and sodic soils.
The proper efficiency in the agricultural production requires the
use of cultivation principles based on the field variability and the
subsequent estimation and mapping of soil properties. The irrigation water quality plays an important role in the agricultural production and has a significant effect on physical and chemical
properties of the soil. Due to the lack of good-quality water specially in arid and semi-arid areas, the risk of soil salinization caused
by the accumulation of salts resulting from the irrigation with lowquality waters increases (Jalali, 2007b). Because of increasing use
of these waters for the irrigation, it is necessary to evaluate their
potentially negative effects on the soil salinity and sodicity and
to implement an appropriate strategy to reduce these problems.
Groundwater is the main source for irrigation of agricultural
fields in Tajarak area located in Hamedan, western Iran. A
geochemical evaluation by Jalali (2007b) indicated that water
samples from wells located in the area were divided into two groups:
(1) relatively low mineralized waters of calcium-bicarbonate
(Ca-HCO3) and sodium bicarbonate (Na-HCO 3) types with total
dissolved solids (TDS) less than 1000 mg L 1, and (2) high mineralized
waters of sodium-sulfate (Na-SO 4) and sodium-chloride (Na-Cl)
types. According to the adjusted sodium adsorption ratio (SAR), the
concentration of Na+ was low in 90% of water samples in group 1, so
they could be used for irrigation without the risk of soil sodicity. However, the severity of the Na + problem in 40% and 37% of
water samples in group 2 was medium and high, respectively.
Therefore, it is necessary to implement a good management to
reduce the sodification of agricultural soils in this region resulting
from the low quality of irrigations waters.
Given the above, the objectives of this study were: (1) the evaluation of spatial dependency and preparation of distribution maps of
the exchangeable sodium percentage (ESP) and related parameters in a selected part of Tajarak area, and (2) the simulation of
long-term effects of saline-sodic and non-saline-sodic irrigation
waters on some chemical properties of a saline-sodic soil under
continuous cultivations of wheat and alternate cultivations of
wheat and sugar beet using PHREEQC as a geochemical computer
program (Parkhurst and Appelo, 1999).

of the area is semi-arid, the average annual precipitation is

317.7 mm, the average annual temperature is 11.4 C, and average
monthly temperatures varies between
2.6 and 24.2 C according
to the data reported by the synoptic weather station of Hamedan
airport by the end of 2010.
A systematic sampling method was performed using grids of
100 m 100 m in an area of 1000 m
300 m (approximately
30 ha). The number of soil sampling points on each x-axis (latitude)
and y-axis (longitude) was 4 and 10, respectively, so, 40 soil samples were taken from vertices of regular square grids. The sampling
depth was 0-30 cm. Latitude and longitude of each sampling point
was recorded by a hand-held global positioning system (GPS)
device. The study area was located between 34.94862
and
34.95679 N latitude equivalent to 321,000 m and 321,300 m in
UTM coordinate and 49.03975 to 49.04284 E longitude equivalent to 3,869,100-3,870,000 m in UTM coordinate. The location of the
sampling points is shown in Fig. 1.
The study area is under conventional tillage and cultivation of
wheat and potato. The salinity and sodicity of soils as a result of
application of low-quality irrigation water and inappropriate irrigation methods are two of major limiting factors in reducing the
soil quality and crop production.
2.2. Analysis of soil samples
The soil samples were air-dried and passed through a 2-mm
sieve after transportation to the laboratory. The particle size distribution was determined using the hydrometer method (Gee and
Bauder, 1986). Soil chemical properties such as pH, electrical
conductivity (EC), and concentration of soluble cations and anions
were measured in the 1:5 soil to water extract and the exchangeable cations were extracted with 1.0 M ammonium acetate
(NH4OAc) solution (pH 7.0). The cation exchange capacity (CEC),
SAR, and ESP were calculated based on measured data (Rowell,
1994).

2. Materials and methods

2.1. Description of the study area
The study area was a part of the agricultural lands with an area
of 30 ha located in Tajarak of Hamedan, western Iran. The climate

Fig. 1. The location of soil sampling with soil sample numbers.

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

2.3. Statistical and geostatistical analysis

Since one of the goals of this study was to investigate the spatial
distribution of ESP and its related parameters, the soil properties
which significantly correlated with ESP were identified using
Pearson correlation coefficients. Prior to the geostatistical analysis,
the distribution of data was tested for the normality using the
Kolmogorov-Smirnov test. All statistical analyses were performed
using the software program SPSS 19.0 for Windows. The analysis
of the spatial dependency was carried out through the calculation
of the semi-variogram function according to the following formula:

ch 2Nh

Zxi Zxi h 2

i1

where c(h) is the semi-variance at a given distance, Z(x i) is the value

of the variable Z at the x i location, and N(h) is the number of pairs of
sample points separated by the lag distance h.
Three major parameters are used to characterize the semivariogram: nugget effect (C 0), sill (C0 + C), and range (A0). The nugget effect represents variations caused by stochastic factors, such
as errors in the measurement or variation that could not be
detected at the minimum sampling distance. Usually, semivariance increases with the lag distance and then approaches a
constant value called the sill. The range indicates the outer limit
at which points in the space still interact spatially and is the distance at which the semi-variogram reaches the sill value. The
semi-variogram of the soil can be described by some theoretical
models including linear, spherical, exponential, and Gaussian models. The selection of the best-fit model of semi-variogram was
made based on the maximum value of coefficient of determination
(R2) and the minimum value of the residual sum of squares (RSS).
The interpolation of data at unsampled locations was performed
using the kriging method and then spatial distribution maps were
prepared. A geostatistical software package (GS+, Gamma Design
Software) was used to analyze the spatial structure of the data
and to define semi-variograms.
2.4. Simulation of long-term effects of applying saline-sodic and nonsaline-sodic waters on some soil chemical properties
In this part, the soil sample with the highest amount of ESP was
used as a representative of saline-sodic soils of study area. The
simulation of long-term changes in some soil chemical properties
including soil solution and exchangeable phases as a result of continuous cultivations of wheat and alternate cultivations of wheat
and sugar beet irrigated with saline-sodic and non-saline-sodic
waters was performed using PHREEQC v.2.18 (Parkhurst and
Appelo, 1999). PHREEQC is a computer program for simulating a
variety of geochemical reactions for a system, including mixing
of waters, addition of net irreversible reactions to solution, dissolving and precipitating phases to achieve equilibrium with the aqueous phase, effects of changing temperature, ion-exchange
equilibria, surface-complexation equilibria, fixed-pressure gasphase equilibria, advective transport, and geochemical inverse
modeling. This program calculates concentrations of elements,
molalities and activities of aqueous species, pH, pe (negative log

303

of the conventional activity of the electron), saturation indices,

and mole transfers of phases to achieve equilibrium as a function
of specified reversible and irreversible geochemical reactions.
The chemical composition of irrigation waters used in the simulation is given in Table 1. Cation selectivity coefficients were calculated using the Gaines-Thomas convention (KGT) in the previous
study (Ranjbar and Jalali, 2015). This equation is based on the
assumption that the cationic mixture on the exchanger phase
behaves as an ideal mixture, i.e., the activity of each exchangeable
species is equal to its equivalent fraction and its activity coefficient
is equal to 1.0. As reported previously (Ranjbar and Jalali, 2015), for
determining KGT values, 25 mL of two types of irrigation water
(Table 1) synthesized in the laboratory were separately added to
2.5 g of the soil sample and after reaching equilibrium, centrifuging, and filtering, the concentrations of Ca2+, magnesium (Mg2+),
K+, and Na+ were measured in supernatants. The exchangeable
cations of the soil samples remained in the centrifuge tubes were
extracted using 25 mL of 1.0 M NH 4OAc solution (pH 7.0). The
activity of soluble cations was obtained through the activity coefficient multiplied by the concentration. The activity coefficients
were calculated using the Debye-Hckel equation as proposed by
the PHREEQC capabilities (Truesdell and Jones, 1974).
The logarithm of KGT values for the exchange of Ca 2+, Mg2+, and K+
over Na+ averaged 0.8, 0.5, and 0.6, respectively (Ranjbar and Jalali,
2015). These values which could successfully simulated
Na+ exchange isotherms in this soil sample in the previous study
(Ranjbar and Jalali, 2015) were used here for the long-term simulation performed by PHREEQC (Table 2).
According to the weather conditions in Hamedan, wheat and
sugar beet are cultivated in early fall and spring, respectively.
The optimal amount of water use by drip-irrigated sugar beet is
7301 m3 ha 1 in this area. Considering this point and the average
annual precipitation equivalent to 3177 m 3 ha 1, the number of
irrigations and precipitation events in every sugar beet growing
season was 7 and 3, respectively. According to the guide given by
Ministry of Agriculture of Iran, the amount of pure water requirement for growing wheat in Hamedan is 4300 m 3 ha 1 which
1970 m3 ha 1 can be supplied by precipitation events. So, regardless of the irrigation method, the pure irrigation water requirement
for wheat cultivation in Hamedan is 2330 m3 ha 1. With regard to
the ratio of the pure irrigation water requirement to the amount of
water supplied by precipitation events (i.e., 1.2) and the number
of main stages needed for irrigation of wheat (i.e., 6), the number
of precipitation events considered in the simulation was 5. In other
words, it was assumed that when the soil was under the
cultivation of irrigated wheat, the crop water requirement was
provided by 6 irrigations and 5 precipitation events during a
cropping season.
Plants take up more cations than anions, so, in order to maintain
the electrical balance inside and outside of roots, they are forced to
produce organic anions and release protons (Rengel, 2003), hence
the soil alkalinity is reduced. According to the report of QDAFF
(2012), the amounts of Ca, Mg, K, and sulfur (S) taken up by the
irrigated wheat grain (with an average yield of 7 t ha 1) plus the
straw (with an average yield of 2 t ha 1) during a cropping season
are 7.5, 16, 75, and 6 kg ha 1, respectively. The amounts of Ca, Mg,
K, Na, and S taken up by the sugar beet tuber (with an average yield

Table 1
The chemical composition of irrigation waters (Jalali, 2007b) used in simulations.
Irrigation water
Non-saline-sodic
Saline-sodic

pH
6.50
8.76

EC
dS m
0.39
8.90

Ca2+
mM

Mg2+
mM

K+
mM

Na+
mM

SAR
(mM)1/2

Cl
mM

CO2
mM

HCO3
mM

SO2
mM

1.00
11.30

0.35
9.25

0.00
4.54

0.35
63.13

0.30
13.93

0.40
54

0.00
1.50

0.80
11.70

0.10
20.00

304

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

Table 2
Input file in PHREEQC for modeling the long-term effect of the application of saline-sodic irrigation water for the first wheat cropping season.

(continued on next page)

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

Table 2 (continued)

305

306

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

of 50 t ha 1) plus the leaf (with an average yield of 40 t ha 1) were

considered to be 78, 44, 248, 69, and 32 kg ha 1, respectively. In
order to quantify the nutrient uptake by the crop and the reduction of
the alkalinity and consequently the acidification of the soil
solution due to the proton excretion by plant roots, calcium
sulfate (CaSO4), calcium carbonate (CaCO 3), magnesium carbonate
(MgCO3), potassium chloride
(KCl), potassium bicarbonate
(KHCO3), and sodium bicarbonate (NaHCO 3) were numerically
removed from irrigation waters. To convert the amount of nutrient
uptake (NU) by the crop in kg ha 1 to that in mmol L 1, the follow-

m ha

10

NU kg ha

m h

105

hs

where m is the atomic mass of the element (g mol 1), h is the soil
depth (cm) and hs is the soil porosity. The soil depth was considered to
be 30 cm. The porosity is defined as follows (Danielson and
Sutherland, 1986):

qb
hs 1

qs

where qb and qs are the average soil bulk density and particle density, respectively. The values of qb and qs were 1.31
and
2.65 g cm 3, respectively, so the soil porosity was equivalent to
0.51.
The seasonal nutrient uptake by crops (mmol L 1) was divided by
the number of irrigations considered in wheat and sugar beet
cropping seasons, equivalent to 6 and 7, respectively and removed
numerically from the irrigation water composition using the
REACTION keyword.
Since non-saline-sodic water was K + free (Table 1), potassium
sulfate fertilizer was added to the composition of irrigation water in
order to compensate the nutritional requirement of crops.
However, due to the presence of sufficient K + in the composition of
saline-sodic irrigation water to meet the requirement of both crops,
the addition of excess K+ to the composition of this type of irrigation
water was ignored.
The mix of precipitation or irrigation water and the soil solution
in the root zone was performed in their respective proportions of
0.5 and 0.5 for wheat cropping seasons and 0.7 and 0.3 for sugar
beet cropping seasons. In order to determine these ratios, the
leaching fraction (LF) was calculated using the following equation
(Oster et al., 1984):

LF Dd
Di

where Dd and Di are the depth of drainage and irrigation waters,

respectively. These parameters were measured as follows: A leaching column with specific cross-sectional area and height was
packed with the soil sample close to bulk density and porosity values mentioned above. Then, a specified volume (equivalent to 2
pore volumes) of saline-sodic irrigation water was added to the soil
column and drainage water was collected. Since the cross-sectional
area at the top and bottom of the soil column was the same, the volume of drainage water divided by the volume of applied water
(equivalent to 0.5) was considered as the leaching fraction. Then,
the amount of water that should be added to the soil to compensate
the leaching fraction and also to provide the pure water requirement of wheat was determined using the following equation
(Oster et al., 1984) by assuming that the irrigation efficiency was
100%:

Viw Vpwr
1 LF

106
Viw L Viw

ing formula was used:

NU mmol L 1

requirement of wheat in the study area (i.e., 4300 m3 ha 1). The

amount of Viw calculated by Eq. (5) was then divided by 11 (i.e., 6
irrigations +5 precipitation events) to obtain the volume of water
needed to be added at each mixing step in every wheat cropping
season. As the simulation was performed for 1 L of the soil solution
in the root zone, the Viw in m3 ha 1 was converted to that in liter per
area which its pore volume was equivalent to 1 L by using the
following equation:

where Viw is the volume of water needed to be applied (m 3 ha 1)

and Vpwr is the volume of water needed to provide the pure water

108

PV

hs

where PV is the pore volume (L) and h and hs are the parameters
described in the Eq. (2). By subtracting the amount of Viw (L) from
1, the mixing ratio of the precipitation or irrigation water and the
soil solution at each mixing step was obtained.
As mentioned before, the amount of water use by drip-irrigated
sugar beet in every cropping season (V iw) is 7301 m3 ha 1 in the
study area. So, there was no need to use the Eq. (5) in this case,
but the sum of this value and the average annual precipitation
(i.e., 10,478 m3 ha 1) was placed in the Eq. (6) and then divided
by 10 (i.e., the number of irrigations and precipitation events during a sugar beet cropping season) to achieve the amount of water
(L) needed to be added to the soil solution at each mixing step. By
subtracting this value from 1, the contribution of the soil
solution at each mixing step was determined.
The partial CO2 pressure in irrigation water was considered to
be equal to the atmospheric value (log pCO2 =
3.5 atm), whereas
it was considered to be 10-fold and constant at log pCO2 =
2.5 atm in the soil solution. The soil solution was concentrated at
the end of each cropping season as a result of the evaporation. The
program written in PHREEQC to simulate the long-term effect of the
application of saline-sodic irrigation water on exchange and solution
phases of the soil is given in Table 2. This input file is for the first
season which soil is under the wheat cultivation. By repeating
the mixing steps, the simulation was performed for 20 consecutive
wheat cropping seasons. The mixing ratios for cropping seasons related to alternate cultivations of wheat and sugar
beet were different as explained above.
3. Results and discussion
3.1. Descriptive statistics
Table 3 indicates that ESP was significantly positively correlated
with the clay content, pH, EC, SAR, and CEC, while it had a significant negative correlation with the silt content. Although there is an
inverse relationship between ESP and CEC based on the calculating
equation, the statistical analysis showed that they were positively
correlated. In other words, an increase in the exchangeable Na +
was in line with increasing CEC in soil samples studied. Shifteh
Somee et al. (2011) reported that ESP was significantly positively
correlated with pH and EC. The descriptive statistics of these
parameters are given in Table 4. Skewness and kurtosis coefficients
are used to describe the shape of the data distribution (Duffera
et al., 2007). The distribution of clay and silt contents was negatively skewed, while the other parameters had a positive skewness.
On the other hand, clay, pH, and CEC showed a positive kurtosis,
whereas the others had a negative one. According to values of
the standard deviation (SD), it can be said that the maximum
and minimum variations in data were respectively observed in
ESP and EC. The values of EC measured in the 1:5 soil to water
extract (EC1:5) varied from 0.24 to 0.70 dS m 1 and averaged
0.43 dS m 1. On the basis of empirical equations given by He
et al. (2013), EC1:5 can be converted to EC in the saturated paste
extract (ECe):

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

307

Table 3
Pearson correlation coefficients for some soil properties.

*
**

Variables

Clay

Clay
Sand
Silt
pH
EC
CEC
SAR
ESP

Sand

0.670**
0.197
0.588**
0.497**
0.854**
0.504**
0.554**

Silt

pH

EC

CEC

SAR

ESP

1
0.722**
0.751**
0.707**
0.849**

1
0.720**
0.702**
0.830**

1
0.709**
0.795**

1
0.918**

1
0.595**
0.004
0.099
0.447**
0.045
0.023

1
0.642**
0.407**
0.334*
0.486**
0.569**

Significant at level of 95% (P < 0.05).

Significant at level of 99% (P < 0.01).

Table 4
Descriptive statistics of some soil properties.
Soil properties

Min.

Max.

Mean

Median

SD

CV (%)

Clay (%)
Silt (%)
pH
EC (dS m 1)
SAR (mmol L 1)1/2
CEC (cmolc kg 1)
ESP (%)

21.24
15.28
8.00
0.24
0.16
17.90
0.43

39.60
31.28
9.01
0.70
4.95
28.40
26.62

34.38
24.40
8.38
0.43
1.67
23.26
7.50

34.88
24.56
8.32
0.43
1.25
23.10
3.85

4.20
3.88
0.28
0.12
1.22
2.34
7.74

12.22
15.90
3.32
26.80
73.23
10.07
103.24

ECe 12:49

EC1:5 2 12:97 EC1:5

0:19

ECe 2:17 EC1:5 4:55 if EC1:5 > 0:4 dS m 1

if EC1:5 < 0:4 dS m 1

7
8

Therefore, the values of ECe in soil samples ranged from 2.2 to

6.1 (average 4.45 dS m 1) indicating the study area included both
saline and non-saline soils. A similar status was observed for ESP.
A range of 0.43-26.62% with an average of 7.5% obtained for ESP
showed that both sodic and non-sodic soils were distributed in
the study area. The greater variation in ESP data compared to significantly correlated parameters indicated that ESP can also be
influenced by management factors such as cropping pattern and
irrigation water quality in addition to structural factors including
parent material, topography, and climate. The coefficient of variation (CV) of soil properties can be categorized as follows: low
(<15%), moderate (15-35%), and high (>35%) variation (Wilding,
1985). The values of CV of SAR and ESP were more than 35% indicating the significant spatial variability (Shifteh Somee et al.,
2011). The lowest and highest values of CV were obtained for pH
and ESP, respectively. The similar results have been reported in
previous studies (Cemek et al., 2007; Shifteh Somee et al., 2011).
Although CV is not enough alone to determine the spatial variability, it is the most discriminating factor for describing variability of a
soil property than other parameters such as SD, mean, and median (Xing-Yi et al., 2007). However, the geostatistical analysis is
necessary to determine the spatial dependency of a parameter in
addition to the statistical analysis.
3.2. The geostatistical analysis
The isotropic semi-variograms were used in order to characterize the spatial variability of parameters (Fig. 2). In addition to the
distance, the semi-variogram can be influenced by direction or
angle between the pair of points. An isotropic semi-variogram
can be considered as an average of semi-variograms obtained from
different directions. Since the variations of soil attributes investigated here followed a similar manner in all directions, or in other
words, the semi-variance depended only on the distance between
samples, the isotropic semi-variograms were used to describe and
quantify the spatial dependency of soil parameters (Fig. 2). The
results of the geostatistical analysis of soil properties investigated

Skewness
1.05
0.50
1.00
0.39
0.75
0.14
1.03

Kurtosis
1.98
0.07
0.11
0.53
0.37
0.46
0.13

and the best-fit models of semi-variograms are given in Table 5.

The semi-variograms of soil physical properties including clay
and silt contents were defined by the Gaussian model, CEC could
be expressed by the exponential model, and the best-fit model
for other parameters was spherical. The maximum and minimum
values of C0 were obtained for the clay content and pH, respectively. Shifteh Somee et al. (2011) reported that the nugget effect
for soil physical properties such as sand, clay, and silt contents was
larger than that for chemical properties indicating the spatial
dependency of chemical properties at shorter distances. At the
same time, the large nugget effect can indicate that sampling intervals are not appropriate to evaluate the spatial dependency of
parameter and there is a need to take more samples at smaller distances. Webster (1985) showed that the nugget effect was attributed to the spatial dependency at distances smaller than the
shortest distance between sampling points.
The ratio of the nugget effect to the sill (i.e., C0/(C0 + C)),
expressed as a percentage, is defined as the relative nugget effect
and an index to demonstrate the ratio of the variation caused by
stochastic factors such as experimental errors to the total variation
of the system. A higher value indicates that stochastic factors play
a major role in the variation, while by reducing this ratio, the effect
of structural factors on the variation of parameters increases
(Cambardella et al., 1994). For example, Table 5 indicates that
the relative contributions of stochastic and structural factors
affecting the variation of the clay content were 28.7% and 71.3%,
respectively. The values of C0/(C0 + C) ratio can demonstrate the
degree of the spatial dependency of variables. Cambardella et al.
(1994) classified the strong, moderate, and weak spatial dependency based on the values of C 0/(C0 + C) ratio of <25%, 75-25%,
and >75%, respectively. According to this classification, pH, EC,
CEC, SAR, and ESP were strongly spatially dependent showing that
their variation was mainly controlled by structural or internal factors including climate, parent material, topography, and soil type,
while clay and silt contents had a moderate spatial dependency
(Table 5) indicating that both stochastic and internal factors played
important roles in their variation. The anthropogenic activities
such as cropping system, fertilization, and irrigation management
can weaken the spatial autocorrelation of soil properties, so, a
weak spatial dependency suggests that the external factors are

308

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

Fig. 2. The semivariograms of some soil properties investigated.

predominant in the spatial pattern. Cemek et al. (2007) reported a

strong spatial dependency for pH, EC, and ESP and a moderate one for
CEC.
When the distribution of soil properties is moderately or
strongly spatially correlated, the mean extent of these distributions

is given by the geostatistical range of the semi-variogram (LopezGranados et al., 2002). The values of the range varied from 106 m
for CEC to 1282 m for the clay content which indicated that the
grid spacing (100 m) was adequate for the characterization of the
spatial variability of soil properties. Big differences among ranges

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

309

Table 5
The parameters of isotropic semi-variogram models of soil properties.

*
**

Variable

Best-fit model

Nugget (C0)

Sill (C0 + C)

Range (A0)
(m)

Nugget/sill
(%)

R2

RSS

Clay (%)
Silt (%)
sian
pH
EC (dS m 1)
CEC (cmolc kg 1)
SAR (mmolc L 1)1/2
ESP (%)

Gaussian
Gaus
8.96
Spherical
Spherical
Exponential
Spherical
Spherical

14.20

49.40

1282

28.7

1.874

0.901**

6.61

1.71
0.968**
0.987**
0.826*
0.907*
0.972**

10 2
2.28
3.36
0.44
3.24
6.12

24.95
2.62 10
2.76 10
0.57
10 3
10 3

4
3

525
1.40 10 3
1.49 10 2
5.51
10.38 10 1
2.52

35.9
645
314
106
450
458

1.766
18.7
18.5
10.4
0.10
0.04

**

1
1.754
1.837
1.875
1.666
1.643

10
10

10
10

Significant at the level of 95% (P < 0.05).

Significant at the level of 99% (P < 0.01).

of different soil variables have been already reported in several

other studies (Lopez-Granados et al., 2002; Cemek et al., 2007;
Shifteh Somee et al., 2011). A larger range value suggests that the
observed values of a parameter are influenced by other values of this
parameter over greater distances (Isaaks and Srivastava, 1989).
Samples separated by distances closer than the range are spatially
related, while samples separated by greater distances are not (Liu
et al., 2011). In other words, the clay content had a spatial dependency over 1282 m, whereas CEC was spatially dependent up to 106 m. This indicated that values of the clay content
correlated with each other over greater distances than other soil
properties, e.g., CEC which had a range of 106 m.
The fractal dimension (D), expressed as the log curve of lag and
semi-variogram, is an indicator of the shape (geometry) of fractal
parameters and can be used to characterize the soil spatial variability (Eghball et al., 1995). Table 5 indicates fractal dimensions
calculated for the spatial distribution of selected soil properties.
In spatial data series, large D-values indicate the importance of a
short-range variation, while small D-values reflect the importance
of a long-range variation. When there are few stochastic factors,
good structure, and simple distribution, the D-values are low,
and vice versa. The D-value of soil properties was in the range of
1.643-1.875. The greatest D-value was found in CEC, while the
smallest D-value was obtained for ESP (Table 5). These values indicated that the variation caused by stochastic factors was greater for
CEC than for ESP, thus, leading to a complex structure of CEC and a
simple structure of ESP (Eghball et al., 1995). Chen et al. (2009)
investigated the spatial variability of five soil fertility factors
including pH, organic matter, available nitrogen (N), available
phosphorous (P), and available K and reported D-values in the
range of 1.641-1.951.
The kriging is one of the most common interpolating methods in
geostatistics which can estimate the values by the local
weighted average (Burgess and Webster, 1980). The kriging
interpolation method is used if data have a normal distribution.
According to the Kolmogorov-Smirnov test, EC, CEC, clay, and silt had
a normal distribution, while pH, SAR, and ESP were not
normally distributed. As the best-fit model of variograms was used in
the kriging, non-normally distributed data were logarithmically
transformed before fitting models to semi-variograms. After
interpolation by the kriging, distribution maps of parameters
investigated were drawn using GS + software (Fig. 3).
The highest amount of clay was observed in the northeast of the
study area and this parameter decreased toward the west. Generally, it can be said that the amount of clay in the north and south
was more than in the middle part. The variation in amounts of clay
can be attributed to the parent material and slope. The distribution
pattern of clay was contrasted with that of silt confirming a negative correlation between these two parameters, although the value
of this correlation was not statistically significant (Table 3).
Soils with higher pH values were mainly distributed in the
northeast parts of the study area, while pH of soils decreased when

moving toward the west. The distribution of pH and clay showed

almost an identical pattern. The lowest values of pH belonged to
the soils distributed in the middle part of the study area. The pH
of soils varied from 8.0 to 9.0 in the 1:5 soil to water extract
(Table 4) and it was less than 8.37 in most soils of the study area
(Fig. 3). Soils with the highest values of EC were distributed in
the northeast parts of the area. The value of EC in most soils was
less than 0.42 dS m 1 in the 1:5 soil to water extract or 5.5 dS m 1
in the saturated paste extract. The spatial variability map showed
that the saline soils were mainly distributed in the northeast and
southeast parts of the area and the west and middle parts have
not the salinity problem. The different distribution of the soil salinity in croplands may be attributed to the irrigation water quality
and the rising groundwater mineralization degree (Gu et al.,
2003; Yan et al., 2006). Like clay, pH, and EC, the highest values
of CEC were mainly observed in the northeast of the study area.
The value of CEC in most soils was less than 23.5 cmol c kg 1. Since
SAR and ESP were positively correlated (Table 3), the distribution
pattern of these two parameters was the same. The value of ESP
in most soils was less than 15%. In other words, most soils of the
area were non-sodic and ESP greater than 15% was only observed
in the northeast part and a small portion of the southeast.
The distribution mapping of various parameters in agricultural
soils can be a very useful tool in the proper and efficient management of soil, water, and plants. Using maps obtained, it can be said
that the management factors such as water quality, irrigation
method, planting systems, and cropping patterns should be revised
and modified in parts with high EC and ESP. Therefore, in the second part of this study, the long-term effect of the irrigation water
quality and crop rotation on some chemical properties of a salinesodic soil sample with ECe of 5.9 dS m 1 and ESP of 26.6% was
investigated using a geochemical simulation.
3.3. The simulation of long-term effects of saline-sodic and
non-saline-sodic irrigation waters on some chemical properties of a
saline-sodic soil
Some selected properties of the soil sample no. 36 before the
long-term simulation, are given in Table 6. This sample was a clay
loam calcareous saline-sodic soil which had the greatest ESP
among soil samples collected from the study area. Table 7 indicates
some chemical properties of this soil sample after 20 years of the
continuous cultivation of wheat and alternate cultivation of wheat
and sugar beet irrigated with non-saline-sodic and saline-sodic
waters. It should be noted that the simulation results at this stage
are comparable with the saturated paste extract conditions due to
the calculation based on the pore volume (here, 1 L). Therefore, the
values of pH, EC, SAR, and the concentration of soluble cations and
anions of the soil sample measured in the 1:5 soil to water extract
(Table 6) cannot be compared with those obtained from simulation
results (Table 7). These results showed that pH and EC of the soil
significantly decreased after 20 years of the continuous cultivation

310

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

Fig. 3. The distribution maps of clay (a), silt (b), pH (c), EC (d), CEC (e), SAR (f), and ESP (g) in the investigated area.

of wheat and alternate cultivation of wheat and sugar beet irrigated with non-saline-sodic water. Furthermore, the values of
pH, EC, and concentrations of cations and anions in the solution
phase at the end of simulation were close to the values of these

parameters in the composition of irrigation water. A decrease in

the concentration of Na+ in exchangeable and solution phases of
the soil under alternate cultivations of wheat and sugar beet was
due to the uptake of this cation from the soil by sugar beet. These

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

Table 6
The selected characteristic of the saline-sodic soil sample before the long-term
simulation.
Parameters

Unit

Quantity

Sand
Silt
Clay
Textural class
pH
EC
Soluble Ca2+
Soluble Mg2+
Soluble K+
Soluble Na+
Soluble HCO3
Soluble Cl
Exchangeable Ca2+
Exchangeable Mg2+
Exchangeable K+
Exchangeable Na+
CEC
ESP
SAR
CaCO3

%
%
%
dS m 1
mM
mM
mM
mM
mM
mM
cmolc kg
cmolc kg
cmolc kg
cmolc kg
cmolc kg
%
(mM)1/2
%

39.48
22.92
37.60
Clay loam
9.01
0.62
0.50
0.20
0.40
3.26
1.20
2.80
11.60
6.20
1.50
7.00
26.30
26.60
3.90
15.00

1
1
1
1
1

pH, EC, and soluble cations and anions were determined in the 1:5 soil to water
extract.

results confirmed that a saline-sodic soil can be modified to a nonsaline-sodic one by changing irrigation water quality. In addition, the
application of this type of water led ESP to fall below 5% at the end of
the first cropping season (data not shown).
The simulation results of the application of saline-sodic water
indicated that the composition of the soil solution phase was similar to the chemical composition of irrigation water used for
20 years. Sugar beet is a sodium-tolerant crop and requires Na +
to complete the life cycle. However, due to the high water requirement, the cultivation of this crop cannot be considered as an
appropriate strategy to avoid or to reduce the soil sodification as
a result of the application of sodic waters. Hence, when the application of sodic water is inevitable, crops with lower water requirements or rainfed crops can be used in the crop rotation. The quality
of irrigation water used in this simulation was according to an
investigation by Jalali (2007b) on the well water samples using
for irrigation of agricultural lands of Tajarak. In fact, the minimum
and maximum values for pH, EC, and concentrations of cations and
anions reported by Jalali (2007b) were used as two completely different types of the irrigation water quality in this study. An
insignificant change in ESP after a-20 year application of salinesodic irrigation water showed that the concentration of Na + in this
type of irrigation water which corresponded to the highest concentration of Na+ in well water samples of Tajarak (Jalali, 2007b) was
nearly similar to the value in the composition of water used for the
irrigation of this soil in real conditions. However, the initial EC e of
the soil was less than the final ECe after 20 years of the application

311

of saline-sodic water. So, it can be concluded that the concentrations of other cations and anions in irrigation water used in this
soil in real conditions are less than the values considered in the
simulation. In other words, despite an unremarkable change in
the sodicity of this soil, the salinity problem has significantly been
increased. Furthermore, the results indicated that in all four scenarios considered in the simulation, Ca2+ was the dominant cation
in the exchange complex, whereas K + had the lowest contribution
among exchangeable cations. The comparison of Tables 6 and 7
indicates that the application of both saline-sodic and nonsaline-sodic irrigation waters and both cropping patterns led to
an increase in the exchangeable Ca 2+ but a decrease in the
exchangeable Mg2+. The extent of increment in the exchangeable
Ca2+ was greater after the long-term application of non-salinesodic water. In addition, the exchangeable K + and Na+ declined
sharply after 20 years of the irrigation with non-saline-sodic water
in both cropping patterns. Since the composition of this type water
was K+ free (Table 1), potassium sulfate fertilizer was added only
enough to provide the crop requirement. Therefore, the mix of this
water and the soil solution phase caused a decreased soluble K +
concentration. Hence, in order to establish a balance between
soluble and exchangeable phases, the exchangeable K + was
released to compensate the reduced concentration of this cation
in the solution phase.
The saturation indices of aragonite, calcite, and dolomite after
the application of saline-sodic water were positive indicating the
precipitation of these minerals in the soil due to greater activity
products of calcium, magnesium, bicarbonate, and carbonate ions
in the soil solution than solubility products of these minerals. On
the other hand, because of low concentrations of cations and
anions in the non-saline-sodic irrigation water, the saturation
indices of these minerals and the others which may be present in
arid and semi-arid areas, e.g., gypsum, anhydrite, halite, and sylvite
were all negative representing their dissolution in the soil.
Hammecker et al. (2009) simulated the evolution of the soil solution in irrigated rice soils over 60 cropping cycles using PHREEQC.
They indicated that the long-term application of non-saline-sodic
waters led to a decrease in the concentration of Ca 2+ in the solution
phase as calcite precipitated. So, the exchange complex reacted to
compensate this trend, providing Ca2+ to the soil solution. Consequently, the proportion of exchangeable Ca 2+ decreased and
exchangeable Mg2+ became dominant. However, simulation results
in our study showed that the saturation index of calcite was negative when non-saline-sodic water was applied, so, the exchangeable Ca2+
increased in comparison with the initial value.
Furthermore, the concentration of Ca 2+ in the composition of
non-saline-sodic irrigation water was more likely to be greater
than that in the composition of irrigation water applied in real
conditions.
According to results of this study, it can be said that the quality
of irrigation water in the north east part of the study area is needed
to be changed by replacing or mixing with good-quality ones in

Table 7
Some chemical properties of the saline-sodic soil after 20 years of the cultivation and irrigation with saline-sodic and non-saline-sodic waters.
Soluble cations and anions
Scenariosa
1
2
3
4

pH
6.78
8.27
6.78
8.29

EC
dS m
0.19
8.64
0.13
9.30

Ca2+
mM

Mg2+
mM

0.76
8.90
0.66
9.98

0.30
8.13
0.25
9.02

K+
mM
0.14
4.24
0.10
4.29

Exchangeable cations

Na+
Cl
mM
mM
0.36
60.18
0.15
64.86

SO2
mM

0.36
48.88
0.00
53.38

HCO3
mM

0.07
18.15
0.10
19.90

0.30
11.48
0.30
11.92

SAR
(mM)1/2

Ca2+
cmolc kg

0.35
14.58
0.16
14.88

21.71
12.18
21.98
12.14

Mg2+
cmolc kg
4.24
5.61
4.09
5.55

K+
cmolc kg
0.20
1.82
0.15
1.74

Na+
cmolc kg
0.13
6.68
0.06
6.84

ESP
%
0.50
25.40
0.23
26.03

a
This column indicates the different scenarios considered in simulations as follows: 20 years of (1) the consecutive cultivation of wheat irrigated with non-saline-sodic
water, (2) the consecutive cultivation of wheat irrigated with saline-sodic water, (3) the alternate cultivation of wheat and sugar beet irrigated with non-saline-sodic water, and
(4) the alternate cultivation of wheat and sugar beet irrigated with saline-sodic water.

312

F. Ranjbar, M. Jalali / Computers and Electronics in Agriculture 121 (2016) 301-312

order to reduce the soil sodicity. In addition, agricultural crops

such as sugar beet and potato with high water requirements
should be replaced with those with low water requirements or
with rainfed crops. Therefore, in order to improve the management
of soil, water, and crop, the irrigation water quality and cropping
patterns should be changed according to the distribution pattern
of soil properties.
4. Conclusions
The exchangeable sodium percentage was significantly positively correlated with the clay content, pH, EC, SAR, and CEC, but
significantly negatively correlated with the silt content. The ESP
data were more distributed compared to the effective parameters
indicating that ESP was also influenced by structural factors such
as parent material and topography. The distribution mapping
results showed that some parts of the study area included saline
soils and the severity of the salinity varied in different parts. Furthermore, the distribution maps indicated that most soils in the
study area did not have the sodicity problem. So, it is expected that
the current cropping and irrigation management in these parts
does not lead to the sodification of soils for a long time. However,
the management in those parts including saline-sodic soils should
be revised and modified. The simulation results of using nonsaline-sodic water in a saline-sodic soil over 20 years of the continuous cultivation of wheat and alternate cultivation of wheat
and sugar beet indicated that the change in the irrigation water
quality led to a significant reduction in EC and ESP. Due to the high
water requirement of sugar beet, it is recommended that the cropping pattern should be changed toward the cultivation of crops
with low water requirements or rainfed crops when good-quality
irrigation water is unavailable.
References
Brejda, J.J., Moorman, T.B., Smith, J.L., Karlen, D.L., Allan, D.L., Dao, T.H., 2000.
Distribution and variability of surface soil properties at a regional scale. Soil Sci.
Soc. Am. J. 64, 974-982.
Burgess, T.M., Webster, R., 1980. Optimal interpolation and isorithmic mapping of
soil properties. I. The semivariogram and punctual kriging. Soil Sci. 31, 315-331.
Cambardella, C.A., Moorman, T.B., Novak, J.M., Parkin, T.B., Karlen, D.L., Turco, R.F.,
Konopka, A.E., 1994. Field-scale variability of soil properties in central Iowa
soils. Soil Sci. Soc. Am. J. 58, 1501-1511.
Cemek, B., Gler, M., Kilic, K., Demir, Y., Arslan, H., 2007. Assessment of spatial
variability in some soil properties as related to soil salinity and alkalinity in
Bafra plain in northern Turkey. Environ. Monit. Assess. 124, 223-234.
Chen, H., Shen, Z., Liu, G., Tong, Z., 2009. Spatial variability of soil fertility factors in
the Xiangcheng tobacco planting region, China. Front. Biol. China 4 (3), 350357.
Danielson, R.E., Sutherland, P.L., 1986. Porosity. In: Klute, A. (Ed.), Methods of Soil
Analysis. Part 1. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI, pp. 443461.
Duffera, M., White, J.G., Weisz, R., 2007. Spatial variability of Southeastern US
Coastal Plain soil physical properties: implications for site-specific
management. Geoderma 137, 327-339.
Eghball, B., Binford, G.D., Power, J.F., 1995. Maize temporal yield variability under
long-term manure and fertilizer application: fractal analysis. Soil Sci. Soc. Am. J.
59,1360-1364.
Ersahin, S., 1999. Spatial variability of some physical and chemical soil properties in
an alluvial field. S. U. J. Agric. Fac. 13, 34-41.

Gee, G.W., Bauder, J.W., 1986. Particle-size analysis. In: Klute, A. (Ed.), Methods of
Soil Analysis. Part 1, 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI, pp.
383-411.
Gu, F.X., Zhang, Y.D., Pan, X.L., 2003. Interaction of land use change and spatial and
temporal dynamics of soil moisture and salinity in arid land. Acta Geogr. Sin. 58
(6),845-853.
Hammecker, C., van Asten, P., Marlet, S., Maeght, J., Poss, R., 2009. Simulating the
evolution of soil solutions in irrigated rice soils in the Sahel. Geoderma 150,
129-140.
He, Y., DeSutter, T., Hopkins, D., Jia, X., Wysocki, D.A., 2013. Predicting EC e of the
saturated paste extract from value of EC1:5. Can. J. Soil Sci. 93, 585-594.
Isaaks, E.H., Srivastava, R.M., 1989. An Introduction to Applied Geostatistics. Oxford
Univ Press, New York, 561p.
Jalali, M., 2007a. Spatial variability in potassium release among calcareous soils of
western Iran. Geoderma 140, 42-51.
Jalali, M., 2007b. Salinization of groundwater in arid and semi-arid zones: an
example from Tajarak, western Iran. Environ. Geol. 52, 1133-1149.
Lee, B.D., Graham, R.C., Laurent, T.E., Amrhein, C., Creasy, R.M., 2001. Spatial
distributions of soil chemical conditions in a serpentinitic wetland and
surrounding landscape. Soil Sci. Soc. Am. J. 65, 1183-1196.
Liu, G., Yang, X., 2008. Spatial variability analysis of soil properties within a field. In:
Li, D. (Ed.), Computer and Computing Technologies in Agriculture. China, pp.
1341-1344.
Liu, W., Su, Y., Yang, R., Yang, Q., Fan, G., 2011. Temporal and spatial variability of
soil organic matter and total nitrogen in a typical oasis cropland ecosystem in
arid region of Northwest China. Environ. Earth Sci. 64, 2247-2257.
Lopez-Granados, F., Jurado-Exposito, M., Atenciano, S., Garcia-Ferrer, A., Sanchez de
la Orden, M., Garcia-Torres, L., 2002. Spatial variability of agricultural soil
parameters in southern Spain. Plant Soil 246, 97-105.
Newman, S., Reddy, K.R., DeBusk, W.F., Wang, Y., Shih, G., Fisher, M.M., 1997. Spatial
distribution of soil nutrients in a northern Everglades Marsh: water
conservation area. Soil Sci. Soc. Am. J. 61, 1275-1283.
Oster, J.D., van Schilfgaarde, J., Russso, D., 1984. Irrigation management and field
salt balance. In: Shainberg, I., Shalhevet, J. (Eds.), Soil Salinity Under Irrigation:
Processes and Management. Springer, Berlin Heidelberg, pp. 175-219.
Parkhurst, D.L., Appelo, C.A.J., 1999. Users Guide to PHREEQC (Version 2): A
Computer Program for Speciation, Batch-Reaction, One-Dimensional Transport,
and Inverse Geochemical Calculations. United States Geological Survey, Water
Resources Investigations Report 99-4259, Washington, DC, 326p.
Parkin, T.B., 1993. Spatial variability of microbial processes in soil: a review. J.
Environ. Qual. 22, 409-417.
QDAFF (Queensland Department of Agriculture, Fisheries, and Forestry), 2012.
Plants, Field crops and pastures, Broadacre field crops, Wheat production,
Wheat nutrition.
Ranjbar, F., Jalali, M., 2015. The effect of chemical and organic amendments on
sodium exchange equilibria in a calcareous sodic soil. Environ. Monit. Assess.
187 (11),683. http://dx.doi.org/10.1007/s10661-015-4894-7.
Rao, P.S.C., Wagenet, R.J., 1985. Spatial variability of pesticides in field soils:
methods for data analysis and consequences. Weed Sci. 33, 18-24. Rengel, Z.,
2003. Handbook of Soil Acidity. CRC Press, New York, 512p. Rowell, D.L., 1994. Soil
Science: Methods and Applications. Longman Group,
Harlow, 360p.
Shifteh Somee, B., Hassanpour, F., Ezani, A., Miremadi, S.R., Tabari, H., 2011.
Investigation of spatial variability and pattern analysis of soil properties in the
northwest of Iran. Environ. Earth Sci. 64, 1849-1864.
Truesdell, A.H., Jones, B.F., 1974. WATEQ, A computer program for calculating
chemical equilibria of natural waters. J. Res. US Geol. Surv. 2, 233-274.
Webster, R., 1985. Quantitative spatial analysis of soil in the field. Adv. Soil Sci. 3, 170.
Wei, Y.C., Bai, Y.L., Jin, J.Y., Zhang, F., Zhang, L.P., Liu, X.Q., 2009. Spatial variability of
soil chemical properties in the reclaiming marine foreland to Yellow Sea of
China. Agric. Sci. China 8 (9), 1103-1111.
Wilding, L.P., 1985. Spatial variability: its documentation, accommodation, and
implication to soil surveys. In: Nielsen, D.R., Bouma, J. (Eds.), Soil Spatial
Variability. Pudoc, Wageningen, pp. 166-194.
Xing-Yi, Z., Yue-Yu, S., Xu-Dong, Z., Kai, M., Herbert, S.J., 2007. Spatial variability of
nutrient properties in black soil of northeast China. Pedosphere 17 (1), 19-29.
Yan, J.F., Chen, X., Luo, G.P., Guo, Q.J., 2006. Temporal and spatial variability
response of groundwater level to land use/land cover change in oases of arid
areas. Chin. Sci. Bull. 51 (1), 51-59.