Euro-Mediterranean Journal for Environmental Integration (2019) 4:40

https://doi.org/10.1007/s41207-019-0130-0

ORIGINAL PAPER

Soil and water salinity evaluation in new agriculture land under arid

climate, the case of the Hassi Miloud area, Algeria
Ahcène Semar1,2 · Tarik Hartani2,3 · Hakim Bachir4 

Received: 25 December 2018 / Accepted: 21 October 2019

© Springer Nature Switzerland AG 2019

Abstract
The food security challenge requires exploring new agricultural lands almost belonging to the arid and Saharian climate
zones. An experimental trial was carried out in 2011 in southern Algeria (Hassi Miloud) where agricultural practices started
a few years ago. Investigations of the soil at different depths (vadose and saturated horizon) and waters sampled from two
irrigations bore holes were made. Irrigation water and subsoil water solution salinity values were close (EC = 3.07 and
EC = 3.16 dS/m, respectively) but the saturated horizon salinity was clearly higher (EC = 9.11 dS/m). The saturation index
and concentration factor were defined to predict possible trends of each type of water. Irrigation waters were undersaturated
in gypsum, anhydrite and halite and close to equilibrium in aragonite, calcite and dolomite. In the vadose and saturated
horizon, the waters were close to the oversaturation state. A correlation matrix was developed to analyze ionic interactions
between the solutions from the vadose and saturated horizon. An important exchange between the two horizons was observed
involving mainly sodium, chlorides and sulfates. We conclude that the use of these waters requires more attention to sustain
agricultural development because they require processes never observed before in dry areas.

Keywords  Agriculture · Exchanges · Aridity · Salts · Subsoil · Water

Introduction on groundwater resources pumped from the complex termi-

nal aquifer (Kuper et al. 2016). These waters are particularly
In North Africa, the cultivation of new agricultural areas mineralized because of the presence of sedimentary stones
has become necessary to supply the growing population’s and high evaporative rates (UNESCO 1972; Swezey 1999;
food needs. However, most of these lands are located in dry OSS 2003).
areas where the soil and water quality are poor and in some The effects of irrigation waters on soil properties have
cases have inefficient drainage systems and high water table been fully discussed in the literature (Richards 1954;
positions (Ben Hassine 2005; Hatira et al. 2007; Askri and Gauthier et al. 1987; Boivin et al. 1988; 1989; Ayers and
Bouhlila 2010; Smadhi et al. 2017). In Algeria, the recent Westcot 1994; Subyani 2005; Al-Ghobari 2011). Daoud
development of agriculture in the Sahara is totally dependent and Halitim (1994) demonstrated that salinization with
poor irrigation waters has adverse effects on the physical,
chemical and biologic properties of the soils and leads to
Editorial handling: Marilyne Soubrand, Chief Editor.
an important yield decrease. Saibi et al. (2009) studied
* Hakim Bachir the water table level dynamic in Oued Souf and mapped
akm7.62@hotmail.fr; hakm7.62@gmail.com the mineral composition of the Sahara waters (Saibi et al.
2013). Semar et al. (2013) used a multidimensional sta-
1
Applied Geology Laboratory, National High School tistical approach to assess the chemical properties of the
of Agronomy (ENSA), Algiers, Algeria
Oued Souf aquifer. Many approaches have been developed
2
Agricultural Water Management Laboratory, National High to explain the spatial variability of the salinity, including
School of Agronomy (ENSA), Algiers, Algeria
the geostatistical approach (Bachir et al. 2016; Sylla et al.
3
University Center of Tipaza Morsli Abdallah, 42000 Tipaza, 1995; Bourgault et al. 1996; Utset et al. 1998; Navarro-
Algeria
Pedren et al. 2007; Zheng et al. 2009; Juan et al. 2011;
4
Division of Bioclimatology and Hydraulics Agriculture, Wei et al. 2014; Long et al. 2014). This article assesses
National Institute of Agronomic Research, Algiers, Algeria

13
Vol.:(0123456789)
 40   Page 2 of 14 Euro-Mediterranean Journal for Environmental Integration (2019) 4:40

the salinity of soil and water irrigation in Hassi Miloud, a Soil and water sampling
new agricultural area in the Algerian Sahara where pome-
granate trees, olive trees, alfalfa and barely are cultivated. Sixty-four samples were taken corresponding to 64 holes
This study focuses on two complementary objectives, distributed according to a 30-m grid plan. For these, three
the first being to evaluate the salinity of soils subjected soil depth levels were prospected corresponding to root
to irrigation from two deep wells capturing the complex depths of 0–0.20 m, 0.20–0.40 m and 0.40–0.60 m (Fig. 3).
terminal aquifer. The second part deals with the possible During the trials, a solid layer of gypsum was found at depth.
hydrochemical influence of the water table on soil salinity. This situation led us to work with only 48 samples over the
The adopted approach was to study the salinity of the soil 64 reaching 0.40 m depth and 34 samples over the 64 reach-
layers and aquifer water. Any ionic interactions between ing 0.60 m depth. In another site, 18 water samples were
the unsaturated and saturated zones were evaluated by the considered at different depths (0.50–1.50 m) to describe
correlation matrix. An analysis of the salinity profiles, the water’s mineral composition in the saturated horizon.
salinity spatial distribution and mineral exchanges between The water sampling occurred 2 days after irrigation, and
irrigation waters and the soil horizons enables a discussion all the samples were collected 10 min after the beginning
of possible trends of the subsoil composition, contribut- of pumping.
ing to the debate on the sustainability of the new Saharian Two bore holes (indicated as F1 and F2) were installed
agricultural models. at about 100 m to irrigate the crops in the study plot and
were taken as landmarks with the following coordinates:
(F1: N32° 04,880′ and E005° 18,050′) and (F2: N32° 04,838′
and E005° 18,098′). The main crops are palm dates, pome-
Materials and methods granates, olives, alfalfa and barley, which are all rather salt
tolerant. A furrow irrigation system, called Seguia, is prac-
Study site description ticed in addition to the drip one, whose use is increasing
rapidly across the region. The drainage layout is made of
This study was conducted from 4 to 11 April 2011 in the open ditches at different parts of the plot (Fig. 3).
oriental Erg of the Algerian Sahara. The study was carried
out in a plot in the Hassi Miloud region, located 13.5 km
from the city of Ouargla (district of Ouargla), limited to the Methodology
north by the agglomeration of N’Goussa (Fig. 1). In the early
1990s, a development program for Saharian agriculture was Three cases are considered here to determine the chemical
dedicated by the state to create “irrigated areas” equipped compositions of soils and waters (Table 1).
with bore holes for irrigation.
Our study site, which covers > 4 ha, consists of sandy – The irrigation waters collected from bore holes;
soils cultivated since the early 2000s. The region is char- – The mineral composition of the soil solution;
acterized by a dry climate with mean annual precipitation – The mineral composition of water in the saturated layer
approaching 38 mm. January, October and November are the (0.50 m and 1.50 m depth).
rainiest months; they receive 8 mm. The mean evaporation
during the 1999–2009 period was 3400 mm. The maximum
evaporation of 500 mm occurs in July (Fig. 2). Spring winds Soil parameter analysis
are common and strongly affect the evaporation rate.
The pumps were designed to extract water from the The methodology adopted for soil solution analyses was rec-
complex terminal aquifer, which is divided into two hydro- ommended by Aubert (1978). As a pre-treatment of the soil
geologic storage tanks: the carbonated “high Eocene Creta- samples, we followed the author’s recommendations to dry
ceous” and the loamy sandy “Mio-Pliocene” (Cornet 1964). the soil samples in open air and sieve them to 2 mm. The
Water age is estimated at between 10,000 and 20,000 years soil samples were packaged in plastic bags and analyzed in
according to radiocarbon methodology (Guendouz et al. the soil science laboratory of the Algerian National School
2003; Ould Baba Sy 2005). The plot was characterized by of Agronomy. The soil pH was measured using the poten-
light, predominantly sandy soils with a particulate structure. tiometric method at a 1:25 soil:water ratio. The physical soil
The soils are distinguished by alkaline pH, high salinity and characteristic (sandy texture) did not allow us to determine
good aeration. Also, these soils have very poor organic mat- the saturated past of soil samples easily because of the dif-
ter levels and low biologic activity; the small amount of ficulties encountered related to the identification of the cri-
organic matter present in the soil comes from the organic teria for achieving saturation of the paste (He et al. 2015).
manure applied in palm plantations. In addition, the 1:5 ratio has the advantage of simplicity,

13
Euro-Mediterranean Journal for Environmental Integration (2019) 4:40 Page 3 of 14  40

Fig. 1  Descriptive elements of the study area. a Geographical loca- Hassi Miloud in the Ouargla Valley. d Piezometric map of the phre-
tion of Ouargla in Algeria; b hydrogeologic cross section of the com- atic groundwater aquifer in Ouargla, Ministry of Water Resources,
plex terminal aquifer in Algeria (adapted from Guendouz et al. 2003). National Office for Sanitation 2004)
The profile of this cross section is presented in a. c Location of

13
 40   Page 4 of 14 Euro-Mediterranean Journal for Environmental Integration (2019) 4:40

Fig. 2  Monthly rainfall and

evaporation data at Ouargla rain
gauging station (1999–2009)

Fig. 3  Location of soil samples

and irrigation boreholes

Table 1  Irrigation and soil Item Number of Parameters

solution under experimental samples
conditions
Irrigation water (bore holes) 2 pH, electrical conductivity (EC), ionic balance
Soil solution in the layer (0.4–0.60 m) 34 pH, electrical conductivity (EC), ionic balance
Saturated layer water 18 pH, electrical conductivity (EC), ionic balance

reducing the time and cost compared with saturation paste result of the measurement is adjusted to a temperature of
extracts, and also dissolves more solutes than the saturation 25 °C (Mathieu and Pieltain 2003).
paste extracts, especially for sparingly soluble and stirring The ionic balance concerned the soil solution extracted
salts (He  et al. 2012; Franzen 2007). Therefore, EC is meas- from a 1:5 soil:water ratio. Calcium, sodium and potas-
ured by the electrical method on 1:5 aqueous extract. The sium are determined by flame photometry. Magnesium is

13
Euro-Mediterranean Journal for Environmental Integration (2019) 4:40 Page 5 of 14  40

determined by atomic absorption spectrophotometry. Sul- holes is calculated from the equation of Todd and Mays
fates are calculated by the gravimetric method based on the (2005), according to the relation (3)
deposition of sulfates in the form of barium sulfate by etch-
ing the extract with a barium chloride solution. Chlorides
HT = 2.5 Ca + 4.1 Mg (3)
are evaluated by the Mohr method based on the titration of where ­Ca2+ and ­Mg2+ are expressed in mg/l.
the soil water extract with silver nitrate by adding potas- The SAR is calculated according to the equation proposed
sium chromate ­(K2CrO4) until a brick red color appears. by Richards (1954) and represented as follows:
Bicarbonates and carbonates are quantified by titration with
sulfuric acid in the presence of indicators (phenolphthalein
/[ ]1∕2
SAR = Na (Ca + Mg)∕2 (4)
and methyl orange). The end of the reaction is indicated by
a color change to orange (Rodier 1997). Na+, ­Ca2+ and ­Mg2+ are expressed in meq/l.
Finally, a geostatistical approach using Variowin software
was applied to study the electrical conductivity variability in
Water parameter analysis
the horizon (0–0.20 m) (Pannatier 1996). The soil salinity
was mapped using kriging to estimate the values of the EC
The irrigation waters and the waters collected from the satu-
unsampled locations using the points around it. The kriging
rated layer were analyzed to assess the ionic balance and pH.
estimation is expressed as follows:
The considered cations were calcium, magnesium, sodium
and potassium, and the considered anions were chlorides, Nnb
sulfates and bicarbonates. The measurement methods were
∑
Z(x0 ) = 𝜆i Z(xi ) (5)
similar to those of the soil solution and were described pre- i=1
viously. Accordingly, the ionic balance analysis is assumed Here z(x0) is the estimator of the mean Z on x0. Z (xi) is
satisfactory when the relative error is < 10%. the known value Z at the point xi. Nnb is the number of data
A water concentration factor (Fc) in a given horizon is points used for estimation, and λi is the kriging weights,
defined by: which are estimated as the solution of the kriging system.
Actual concentration ion in chlorides The weightings involved in the linear combination are
Fc (meq/l) =
The minimum chlorides concentration in all samples obtained by solving the minimization problem whose equa-
(1) tions depend on the theoretical variogram and the geometric
As the chlorides are not degraded in the environment and configuration of the EC data point’s knowledge. The equa-
tend to remain in solution once dissolved, the concentration tion of the semi-variogram is expressed as:
factor is a good indicator of the exchanges between the sub- m
soil horizons. 1 ∑{ }2
𝛾(h) = Z(Xi + h) (6)
In addition, “Diagram” software was run to determine the 2m i
water properties and calculate mineral saturation indexes
where h is the distance between Xi and Xj; m is the number
(Simler 2007, Parkhurst and Appelo 2013). The saturation
of pairs that are separated by the distance h. The model vali-
index (SI) is calculated using the following relationship:
dation is completed by calculating the indicative goodness
PAI of fit (IGF).
SI = log (2)
Ks
where PAI is the activity product of the ions, and Ks is the
solubility product of the considered mineral. Results
The equilibrium state between a given mineral and the
water is reached when the saturation index (SI) = 0. SI < 0 Chemical characteristics of the waters
indicates the soil solution is undersaturated, and SI > 0
indicates oversaturation. Consequently, a given solution is The comparison of the two bore holes’ waters is shown in
undersaturated when it moves to dissolution. Contrarily, Table 2. The pH is neutral to slightly alkaline and equals
oversaturation occurs when the solution moves to precipita- 7.9 and 7.5 in F1 and F2, respectively. The salinity in F1
tion (Deutsch and Siegel 1997). Two important water quality waters is lower than that of F2 with an EC value of 1.84
parameters are used in the chemical assessment of water: and 3.07 dS/m, respectively. In addition, all the mineral
the water hardness and sodium adsorption ratio (SAR). The concentrations of F2 waters are higher than those of F1
water hardness expressed in mg/l of ­CaCO3 in F1 and F2 because sodium and sulfates are the major cations and ani-
ons, respectively, observed in waters. We conclude that

13
 40   Page 6 of 14 Euro-Mediterranean Journal for Environmental Integration (2019) 4:40

Table 2  Bore holes (F1 and F2) pH CE Ca++ Mg++ Na+ K+ Cl− SO42− HCO3− SAR Hardness
analysis results
F1 7.9 1.84 10.27 2.38 19.43 0.46 14.8 18.51 1.18 7.74 632.5
F2 7.5 3.07 14.22 6.41 30 0.74 21.8 25.3 3.39 9.34 1031.5

EC in dS/m at 25 °C, ions in meq/l and hardness in mg/l of C

­ aCO3

Table 3  Saturation index (SI) Anhydrite Aragonite Calcite Dolomite Gypsum Halite

values
F1 − 0.82 0.15 0.29 0.06 − 0.60 − 5.29
F2 − 0.67 0.29 0.43 0.64 − 0.45 − 4.96

Fig. 4  Frequency of the mineral 250

contents in the saturated zone Ca⁺⁺ Mg⁺⁺ Na⁺ K⁺ Cl¯ SO₄¯ ¯
200
Concentra on (meq/l)

150

100

50

0
4 5 6 7 8 11 15 22 26 27 38 39 45 47 48 50 51 52
Water samples

the waters’ chemical facies is dominated by sodium sul- mean values of 82.09 meq/l and 124.35 meq/l, respectively
fate ­(NaSO4) with ionic concentrations ranking as follows: (Fig. 4).
­Na+ > Ca2+ > Mg2+ > K+ and ­SO42− > Cl− > HCO3−. In the saturated horizon, the salinity of the solution is also
The depiction of the mineral composition of the waters dominated by sodium sulfate and confirms the conclusions of
in a Riverside chart exhibits high salinity and a medium Semar et al. (2013). The Piper’s chart shows that the cation
alkaline trend for F1 defined by the class (C3-S2). For F2, amount decreases as ­Na+ > Ca2+ > Mg2+ > K+ and that the
a very high salinity and high alkaline trend exist, defined anion amount decreases as ­SO42− > Cl− > HCO3− (Fig. 5).
by the class (C4-S3). Irrigation with these waters should be In some solutions collected from the saturated horizon
carefully practiced particularly in case of F2 waters because (6 samples), the magnesium amount was higher than that of
the chloride content exceeds 10 meq/l (Ayers and Westcott calcium. This observation occurred indistinctly when F1 and
1994). F2 waters were used in irrigation. According to Nezli et al.
The saturation index (SI) indicates an oversaturation close (2007), the sodium sulfate trend encountered in only 7.28%
to equilibrium in aragonite, calcite and dolomite, while it of the region’s water enables an ionic exchange between
indicates an undersaturation in gypsum, anhydrite and halite sodium and magnesium. The concentration values of the
(Table 3). waters vary from 1.20 and 2.80; five of the values were >
2.00. This is because of a gypsum layer that reduces the ver-
Saturated horizon salinity tical drainage and limits the chloride transfer in deep levels.
The obtained saturation indexes (SIs) indicate that gyp-
The pH of the collected solution ranged from neutral to sum and anhydrite contents vary in an opposite way to
slightly alkaline (7.30–8.10). A high level of salinity was vadose (Fig. 6). An overestimation of aragonite, calcite
observed, and EC ranged between 7.05 and 11.65 dS/m with and dolomite contents (SI values vary between 0 and 2) is
a variation coefficient of 16.14% and mean of 9.11 dS/m. observed for samples 4, 7, 8, 22, 27, 38, 39, 45 and 50. Nezli
Major cations and anions were sodium and sulfates with et al. (2007) have also demonstrated that these solutions are
oversaturated with carbonates and that the concentration fac-
tor is such that log (Fc) < 0.20. An underestimation of the

13
Euro-Mediterranean Journal for Environmental Integration (2019) 4:40 Page 7 of 14  40

Fig. 5  Piper diagram of the

saturated horizon’s water

Fig. 6  Saturation index varia-

tion in the saturated zone

halite content was also reported by these authors indicating the water belongs to the C5-S4 class (Fig. 7). These values
SI values between − 4.5 and − 5.5 (Fig. 6). express a strong saline and alkaline trend due to the continu-
Concerning the water solution in the saturated horizon, ous interaction among irrigation water leaching, capillary
the SAR index varies from 9.84 to 25.22, which implies that ascension and water table rising. In this horizon, the chloride
concentration exceeds the 10 meq/l threshold and ranges

13
 40   Page 8 of 14 Euro-Mediterranean Journal for Environmental Integration (2019) 4:40

Soil salinity in the 0.40–0.60 m layer

The soil salinity analysis included 38 samples. The pH of

the soil solutions varied from neutral (7.1) to slightly alka-
line (7.9). The average electrical conductivity of the soil
reached 3.16 dS/m with a variation coefficient of 34%. The
main anions present were sulfates and chlorides, and a low
concentration of bicarbonate was detected. The main cations
present were calcium and magnesium (Fig. 10).
Chemical profiles were dominated by sulfate and sodium
with the following ranking: ­Na+ > Ca2+ > Mg2+ > K+ and
­SO42− > Cl− > HCO3− (Fig. 11).

Interactions between the vadose and saturated

horizons

The hydrochemical interactions between the ions of the satu-

rated and unsaturated horizons were assessed using a cor-
Fig. 7  Riverside chart of groundwater samples relation matrix (Table 4). To focus on this point, the ionic
concentrations in the layers (0.40–0.60 m) and saturated
zone water were determined. High negative coefficients were
between 11.09 and 30.59 meq/l. This salinity trend is prob- obtained particularly between chlorides (­ Cl−) and sulfates
ably enhanced by the furrow irrigation practice, poor surface ­(SO42−) and between sodium ­(Na+) and potassium ­(K+)
drainage performance and solid gypsum layer. (r = − 0.65 and r = − 0.56, respectively).
In the saturated horizon, high positive correlations were
Soils observed in the following cases: ­(Na+–SO42−), (CE–SO42−),
(CE–Na+) and (­ Mg2+–SO42−) with correlation coefficients
Four types of salinity profiles were observed in Hassi Mil- of 0.84, 0.73, 0.58 and 0.49, respectively. Semar et  al.
oud: rising, falling, convex and concave. The rising profile (2013) have already observed a strong correlation between
was the major one observed (17 cases). The average salin- ­SO42− and M ­ g2+ (r = 0.59) in Oued M’ya, and they dem-
ity of the 0.40–0.60 m layer was 3 dS/m and increased to onstrated that the ­Ca2+ and ­SO42− (r = 0.87) correlation
7.4 dS/m in the 0–0.20 m layer (Fig. 8a). The decreas- explains 22% of the variance of the (F2) factorial axis. The
ing profile showed salinity increasing from 4.5 dS/m at sulfates were the second major component after chlorides
the surface to 6.3 dS/m at deep layers (Fig. 8b). The EC in the water table. High correlation coefficients illustrate
of the convex salinity profiles was high at the horizon the strong relationships between the vadose and saturated
(0.20–0.40 m), approaching 4.9 dS/m, but in the deeper horizons: magnesium (r = 1), bicarbonates (r = 1), potas-
layers it reached 4 dS/m and 3.7 dS/m (Fig. 8c). The con- sium (r = 0.98) and calcium (r = 0.95). This means that an
cave salinity profiles were illustrated by eight cases, and intensive exchange takes place between the two horizons;
their average salinity was 3.6 dS/m, 5.3 dS/m and 4.9 dS/m in addition, the exchanges concerning sodium (r = 0.76),
in the 0.20–0.40 m, 0–0.20 m and 0.40–0.60 m layers, chlorides (r = 0.55) and sulfates (r = 0.51) are relatively
respectively (Fig. 8d). important. Electrical conductivity (EC) was mostly linked
to sulfate and sodium concentrations (r = 0.70 and r = 0.62,
Soil texture and salinity in the 0–0.20 m layer respectively). The chloride and sulfate correlation in one
part and potassium and sodium correlation in another were
According to the soil’s granular analysis, the 0–0.20 m layer strong but had a negative value (r = − 0.65 and r = − 0.55,
was sandy, comprising 57% thin and 29.4% thick sand. The respectively). This means that the concentrations varied in
fraction of clay combined with loam was nearly 13.6%. The an opposite way. Contrarily, the concentrations of sulfates
mapping of EC for this horizon, where major crop roots and sodium on one hand and magnesium and sulfates on the
grow, showed values ranging between 3 and 8 dS/m (Fig. 9). other varied in a similar way.
This is due to the water quality of each borehole, the irriga-
tion system and the soil topography.

13
Euro-Mediterranean Journal for Environmental Integration (2019) 4:40 Page 9 of 14  40

Fig. 8  Salinity pattern of profiles

13
 40   Page 10 of 14 Euro-Mediterranean Journal for Environmental Integration (2019) 4:40

silts accounted for only about 14%. The kriging salinity

map of the surface layer varied from 3 to 8 dS/m, marked
by salinity ranges closely related to the gravity irrigation
system and the quality of the irrigation water. Experiments
conducted in two sprinkler irrigation fields in the Algerian
Sahara showed that salinity, which was initially < 2 dS/m
in the surface horizons, increased to > 12  dS/m after
five irrigation campaigns using the waters of the Albian
groundwater with an EC of 8.5 dS/m (Djili et al. 2003).
Furthermore, the ascending saline profile type represented
50% of the other profiles. This indicated a phenomenon of
soil salinization in progress. This salinization is connected
to the capillary rise of the waters of the shallow saline
layer combined with the excessive evaporation. In addi-
tion, saline efflorescence appeared at the soil surface level.
This is explained by the effect of the capillary rise rein-
forced by the evaporative capacity of the atmosphere. The
descending profile (12%) underlines a possible leaching of
the soluble salts and the presence of a deep gypseous crust,
probably related to the presence of a solid gypsum layer in
depth and to the salt leaching after many irrigation stages.
The concave saline profiles (23%) seem likely to corre-
Fig. 9  Mapping of EC (dS/m) by krigeage method in the 0–0.20  m spond to the resumption of the ascending process of salini-
layer zation after a period of desalination. The convex types
(15%) showed a probable temporary desalinization whose
saline dynamics were characterized by a capillary rise and
Discussion a lixiviation of the salts. As a reminder, the water sam-
ples were taken 2 days after the irrigation cycle. The EC
The Ouargla rain gauging station showed irregular precipi- kriging map of the surface layer (0–20 cm) indicates high
tation behavior. The rainy months were January, October salinity up to 8 dS/m in places. Research conducted in the
and November, with 8 mm each. During the other months, oasis environment, in the irrigated perimeter of Segdoud
the rainfall did not exceed 4  mm. On the annual scale situated in southwest Tunisia, shows that between winter
(1999–2009), 38 mm fell. Evaporation related to precipi- 1994 and spring 1995, the average salt concentration on
tation was intense—the annual average was on the order of the surface horizon increased from 7 to 8 dS/m (Askri and
3400 mm. The minimum value (100 mm) was recorded in Bouhlila 2010). In the Tunisian case, the irrigation water
January and the maximum (500 mm) in July. This evapora- is pumped from the intercalar continental with an average
tion was accentuated by winds. The study site soils were EC of 5.2 dS/m. These results may lead to a reduction in
very sandy (86%) with a sand content almost double (57%) yields because the tolerance threshold to the concentration
that of coarse sand (29%). However, the rate of clay and of salt in the root zone is specific to each crop (Hamdia

Fig. 10  Ion content (meq/l) of 80

soil solutions in the 0.40–
Ca⁺⁺ Mg⁺⁺ Na⁺ K⁺ Cl ˉ SO₄ ˉ ˉ
70
0.60 m layer
Concentration (meq/l)

60

50

40

30

20

10

0
3 4 5 6 7 8 11 12 14 15 16 17 18 19 20 22 25 26 27 28 29 34 38 39 40 41 45 47 48 50 51 52 60 63
Soil samples

13
Euro-Mediterranean Journal for Environmental Integration (2019) 4:40 Page 11 of 14  40

Fig. 11  Piper chart of ion contents (meq/l) in the soil solutions

and Shaddad 2010). As a result, salt-resistant crops are bloedite and others) appears to be a distinctive feature of
suggested, and salt leaching of the drainage system is rec- the hyper-desert soil. In this environment, they are more
ommended in this case. This is related to the combined stable than simple sulfates. The waters of the saturated
effects of irrigation and upwelling from the saturated zone. zone had a ­SO42− and ­Na+-controlled EC and revealed the
The chemical analysis of the irrigation water of two bore- same facies as before; they belong to the C4-S4 class with
holes (F1 and F2) capturing the terminal complex of the chloride contents > 10 meq/l. The chemical exchanges
Hassi Miloud perimeter indicated a slightly alkaline pH, between the unsaturated and saturated zones were illus-
the ECs were close to 1.8 and 3.1 dS/m, respectively, and trated by the ions ­Mg2+, ­HCO3−, ­K+, ­Ca2+, ­Na+, ­Cl− and
the chemical facies was sulfated sodium. Considering the ­SO42−.
EC values present at the two boreholes (F1 and F2), to Once irrigation waters, soil solutions of the vadose and
mitigate the risk of soil salinization, the use of F1 in crop saturated horizons had been assessed according to various
irrigation is recommended as the EC value was relatively classification charts, we found that the irrigation waters
lower (1.84 dS/m) than that of F2 (3.07 dS/m). belong to the C3-S2 and C4-S3 classes according to the
The irrigation classes defined by the Riverside dia- Riverside classification. Salinity, alkalinity and excess chlo-
gram were C3-S2 and C4-S3, respectively, for F1 and F2 ride were the main characteristics of the F2 well’s water.
and show that the waters pumped from the second water High salinity levels were observed in the vadose (0–0.20 m),
point represented a danger of salinization and alkalization. and primary ions were sodium and sulfates. In the horizon
However, research in Tunisia showed that medium saline (0.40–0.60  m), the solution properties belonged to the
water can be used for irrigation without major difficulties C4-S4 class because of a significant chloride concentration
(Cointepas 1964). Chloride levels exceed the allowable (> 10 meq/l).
limit of toxicity for plants usually grown in intercrops.
Drilling water (F2) tended to deposit carbonate minerals
(aragonite, calcite and dolomite), unlike gypsum and hal- Conclusion
ite, which dissolve. As for drilling water (F1), only calcite
tended to settle; the other minerals were either in equilib- The study conducted in the Hassi Miloud (Ouargla) region
rium or undersaturated. The soil solution of the deep layer was a representative case for many areas with an arid climate
(40–60 cm) had an EC on the order of 3 dS/m with a sul- irrigated by relatively mineralized waters of the complex
fated sodium chemical facies. According to Hamdi-Aissa terminal and located near a shallow-surface water table. The
et al. (2004), the presence of mixed sulfates (glauberite, chemical analysis of the irrigation water of two boreholes

13
 40  

13
Page 12 of 14

Table 4  Correlation matrix between soil solutions (VZ) in the 0.40–0.60 m layer and saturated horizon (SZ)
pHvz Cevz Cavz Mgvz Navz Kvz Clvz SO4vz HCO3vz pHsz Cesz Casz Mgsz Nasz Ksz Clsz SO4sz HCO3sz

pHvz 1
Cevz − 0.30 1
Cavz 0.33 − 0.36 1
Mgvz 0.28 − 0.36 0.40 1
Navz − 0.34 0.35 − 0.10 − 0.18 1
Kvz 0.46 − 0.24 0.22 0.34 − 0.56 1
Clvz 0.003 0.14 − 0.13 0.16 − 0.19 − 0.05 1
SO4vz − 0.10 − 0.13 0.42 0.08 0.46 − 0.12 − 0.65 1
HCO3vz 0.18 0.14 0.37 − 0.09 0.09 − 0.02 − 0.33 0.43 1
pHsz − 0.20 − 0.33 0.24 0.22 0.01 0.11 − 0.11 0.07 − 0.09 1
Cesz − 0.05 0.01 0.42 0.27 0.62 − 0.34 − 0.41 0.70 0.46 0.08 1
Casz 0.27 − 0.44 0.95 0.30 − 0.22 0.24 − 0.23 0.40 0.30 0.36 0.32 1
Mgsz 0.29 − 0.36 0.40 1.00 − 0.18 0.34 0.16 0.08 − 0.08 0.22 0.27 0.30 1
Nasz − 0.14 0.001 0.10 0.19 0.76 − 0.55 0.07 0.25 − 0.17 0.10 0.58 − 0.03 0.19 1
Ksz 0.41 − 0.26 0.30 0.36 − 0.46 0.98 − 0.05 − 0.07 − 0.07 0.16 − 0.27 0.31 0.36 − 0.40 1
Clsz 0.35 − 0.34 0.34 0.44 − 0.27 0.25 0.55 − 0.22 0.04 − 0.06 0.02 0.27 0.44 0.15 0.28 1
SO4sz − 0.07 − 0.14 0.44 0.49 0.60 − 0.31 − 0.16 0.51 0.01 0.29 0.73 0.35 0.49 0.84 − 0.17 0.09 1
HCO3sz 0.18 0.15 0.36 − 0.10 0.09 − 0.01 − 0.33 0.43 0.99 − 0.08 0.45 0.29 − 0.09 − 0.19 − 0.07 0.02 − 0.01 1

In bold, significant values at 0.05 level

Euro-Mediterranean Journal for Environmental Integration
(2019) 4:40
Euro-Mediterranean Journal for Environmental Integration (2019) 4:40 Page 13 of 14  40

(F1 and F2) capturing the complex terminal of the Hassi Boivin P, Hachicha M, Job JO, Loyer JY (1989) Une méthode de car-
Miloud perimeter indicated a slightly alkaline pH. The ECs tographie de la salinité des sols: conductivité électromagnétique
et interpolation par krigeage. Sci Sol 27:69–72
were close to 1.8 and 3.1 dS/m, respectively, and the chemi- Bourgault G, Journel AG, Rhoades JD, Corwin DL, Lesch SM (1996)
cal facies was sodium sulfated. In this article, the salinity Geostatistical analysis of a soil salinity data set. Adv Agron
map of the surface layer showed variability ranging from 58:241–292
3 to 8 dS/m marked by salinity ranges closely related to Cointepas JP (1964) Résultats d’un essai d’irrigation à l’eau salée
en Tunisie. In: 8th Inter. Congress of soil science, Bucharest,
the gravity irrigation system and the quality of the irriga- Romania, pp 793–802
tion water. Saline profiles were dominated by ascending and Cornet A (1964) Introduction à l’hydrogéologie du Sahara, Algérie.
concavo-convex types. They were related to the combined Revue Géographie Physique et Géologie Dynamique 6:5–72
effects of irrigation and upwelling from the saturated zone. Daoud Y, Halitim A (1994) Irrigation et salinisation au Sahara algé-
rien. Sécheresse 5:151–160
In the human-modified landscapes of the Algerian Sahara, Deutsch WJ, Siegel R (1997) Groundwater geochemistry: fundamen-
the high mineral content of irrigation waters and soils is a tals and applications to contamination. CRC Press, Boca Raton
critical issue. In the Hassi Miloud case study, most of the Djili K, Daoud Y, Gaouar A, Beldjoudi Z (2003) La salinisation des
cultivated crops grow in the 0–0.60 m horizon, which is sols au Sahara. Conséquences sur la durabilité de l’agriculture
dans les nouveaux périmètres de mise en valeur. Science et
why we focused on the interactions between the subsurface changement planétaires Sécheresse 14(4):241–246
and groundwater mineral properties. The correlation matrix Franzen D (2007) Salt accumulation processes. North Dakota State
highlighted which ions must be considered during irrigation University, Fargo
and enabled predicting trends in mineral salinity exchanges Gauthier TD, Seitz WR, Grant CL (1987) Effects of structural and
compositional variations of dissolved humic materials on pyr-
between the saturated and vadose horizons. Finally, the pro- ene Koc values. Environ Sci Technol 21:243–252
posed approach provides a useful tool for stakeholders and Guendouz A, Moulla AS, Edmunds WM, Zouari K, Shands P,
agricultural managers to adjust their sustainable develop- Mamou A (2003) Hydrogeochemical and isotopic evolution
ment strategies for the same kind of environments. of water in the complex terminal aquifer in Algerian Sahara.
Hydrogeol J 11:483–495
Hamdia MA, Shaddad MAK (2010) Salt tolerance of crop plants. J
Acknowledgements  This review article was not supported by any grant Stress Physiol Biochem 6(3):64–90
or funding. The authors thank Mrs. Benkedidah Fatima Zohra for her Hamdi-Aissa B, Valles V, Aventurier A, Ribolzi O (2004) Soils
contributions of carrying out water and soil analyses and enriching and Brine Geochemistry and Mineralogy of Hyperarid Desert
exchanges related to the topic. Playa, Ouargla Basin, Algerian, Sahara. Arid Land Res Manag
18(2):103–126. https​://doi.org/10.1080/15324​80490​27965​6
Compliance with ethical standards  Hatira A, Baccar L, Grira M, Gallali T (2007) Analyse de sensi-
bilité du système oasien et mesures de sauvegarde de l’oasis de
Conflict of interest  On behalf of all authors, the corresponding author Métouia (Tunisie). J Water Sci 20:59–69
states that there is no conflict of interest. He Y, DeSutter T, Prunty L, Hopkins D, Jia X, Wysocki DA (2012)
Evaluation of 1:5 soil to water extract electrical conductiv-
ity methods. Geoderma. https ​ : //doi.org/10.1016/j.geode​
rma.2012.03.022
He Y, DeSutter TM, Hopkins DG, Wysocki D, Clay DE (2015)
References Relationship between 1:5 soil/water and saturated paste
extract sodium adsorption ratios by three extraction methods.
Soil Sci Soc Am J 79:681–687. https​://doi.org/10.2136/sssaj​
Al-Ghobari HM (2011) Effect of irrigation water quality on soil salin- 2014.09.0384
ity and application uniformity under center pivot systems in arid Juan P, Mateu J, Jordan MM, Mataix-Solera J, Meléndez-Pastor I,
region. J Appl Sci Res 7(7):72–80 Navarro-Pedreño J (2011) Geostatistical methods to identify
Askri B, Bouhlila R (2010) Evolution de la salinité dans une oasis and map spatial variations of soil salinity. J Geochem Explor
moderne de Tunisie. Etude et Gestion des Sols 17:197–212 108:62–72
Aubert G (1978) Méthodes d’analyses des sols, 2ème edn. Centre Kuper M, Faysse N, Hammani A, Hartani T, Marlet S, Hamamouche
régional de Documentation Pédagogique, CRDP Marseille MF, Ameur F (2016) Liberation or Anarchy? The Janus nature
Ayers RS, Westcot DW (1994) Water quality for agriculture: FAO. of groundwater use on North Africa’s new irrigation frontiers.
Irrig Drain 29:1–130 Integrated groundwater management, concepts, approaches
Bachir H, Semar A, Mazari A (2016) Statistical and Geostatisti- and challenges. Springer, Cham, pp 583–615. https​: //doi.
cal analysis related to geographical parameters for spatial and org/10.1007/978-3-319-23576​-9
temporal representation of rainfall in semi-arid environments Long Xiao-Hua, Zhao Jie, Liu Zhao-Pu, Rengel Zed, Liu Ling, Shao
the Case of Algeria. Arab J Geosci 9(7):486–498. https​://doi. Hong-Bo, Tao Ya (2014) Applying geostatistics to determine
org/10.1007/s1251​7-016-2505-8 the soil quality improvement by Jerusalem artichoke in coastal
Ben Hassine H (2005) Effets de la nappe phréatique sur la salinisa- saline zone. Ecol Eng 70:319–326
tion des sols de cinq périmètres irrigués en Tunisie. Etude et Mathieu C, Pieltain F (2003) Analyse chimique des sols—méthodes
Gestion des Sols 12:281–300 choisies. Editions Tec et Doc, Lavoisier
Boivin P, Brunet D, Job JO (1988) Conductivimétrie électromagné- Ministry of Water Resources-National Office of Sanitation (2004)
tique et cartographie automatique des sols salés: une méthode Etudes d’Assainissement des eaux résiduaires, pluviales et
rapide et fiable. Cahiers ORSTOM Série Pédologie 24:39–48 d’irrigation. Mesures complémentaires de lutte contre la
remontée de la nappe phréatique—Vallée de Ouargla. Mission

13
 40   Page 14 of 14 Euro-Mediterranean Journal for Environmental Integration (2019) 4:40

II—Rapport final document référence 6069.01/RN097. Décem- Semar A, Saibi H, Medjerab A (2013) Contribution of multivari-
bre 2004—Etude réalisée par le bureau d’études BG (Bonnard ate statistical techniques in the hydrochemical evaluation of
et Gardel) pour le compte du Ministère des Ressources en Eau groundwater from the Ouargla phreatic aquifer in Algeria. Arab
et Maître d’ouvrage ONA J Geosci Springer 6:3427–3436
Navarro-Pedren J, Jordan MM, Melendez-Pastor I, Gomez I, Juan Simler R (2007) Diagrammes. Laboratoire d’hydrogéologie
P, Mateu J (2007) Estimation of soil salinity in semi-arid land d’Avignon. Université d’Avignon, Logiciel d’hydrochimie mul-
using a geostatistical model. Land Degrad Develop 18:339–353 tilangage en distribution libre. France
Nezli IE, Achour S, Djabri L (2007) Approche géochimique des pro- Smadhi D, Zela L, Bachir H (2017) Droughts in semi-arid cereal
cessus d’acquisition de la salinité des eaux de la nappe phréa- regions of Algeria. J Appl Fundament Sci 09(02):1063–1073.
tique de la basse vallée de l’Oued M’Ya (Ouargla). Larhyss J https​://doi.org/10.4314/jfas.v9i2.29
6:121–134 Subyani AM (2005) Hydrochemical identification and salinity prob-
OSS (2003) Observatoire du Sahara et du Sahel. Système aquifère lem of ground-water in Wadi Yalamlam basin, Western Saudi
du Sahara septentrional. Une conscience de bassin. Hydro- Arabia. J Arid Environ 60:53–66
géologique 2, 2ème Edition, Tunis Swezey C (1999) The lifespan of the Complexe Terminal Aquifer,
Ould Baba Sy M (2005) Recharge et paléorecharge du système Algerian-Tunisian Sahara. J Afr Earth Sci 28:751–756
aquifère du Sahara septentrional. Thèse Faculté des Sciences Sylla M, Stein A, Van Breemen N, Fresco LO (1995) Spatial vari-
de Tunis, Tunisie ability of soil salinity at different scales in the mangrove rice
Pannatier Y (1996) Variowin-software for spatial data analysis in agro-ecosystem in West Africa. Agric Ecosyst Environ 54:1–15
2D. Springer, New York Todd DK, Mays LM (2005) Groundwater hydrology, 3rd edn. Wiley,
Parkhurst DL, Appelo CAJ (2013) Phreeqc version 3—a computer Hoboken
program for speciation, batch–reaction, one dimensional trans- UNESCO (1972) Etude des ressources en eau du Sahara Septen-
port, and inverse geochemical calculations. Geological Survey, trional. Algérie—Tunisie, Rapport sur les résultats du projet:
Denver. https​://doi.org/10.3133/wri99​4259 conclusions et recommandations, Paris
Richards LA (1954) Diagnosis and improvement of saline and alka- Utset A, Ruiz ME, Herrera J, Ponce de Leon D (1998) A geostatis-
line soils, vol 60. USDA Agric Handbook, Washington tical method for soil salinity sample site spacing. Geoderma
Rodier J (1997) L’analyse de l’eau. Eaux naturelles, eaux résiduaires, 86:143–151
eau de mer, 8th edn. Collection Dunod, Paris Wei Hu, Shao Ming An, Wan Li, Si Bing Cheng (2014) Spatial vari-
Saibi H, Semar A, Mesbah M, Ehara S (2009) Variographic analysis ability of soil electrical conductivity in a small watershed on the
of water table data from the Oued Souf phreatic aquifer, north- Loess Plateau of China. Geoderma 230–231:212–220
eastern part of the Algerian Sahara. Arab J Geosci 2:83–93 Zheng Zhong, Zhang Fengrong, Ma Fuyu, Chai Xurong, Zhu Zhan-
Saibi H, Bahri F, Semar A (2013) Hydrochemistry and bacteriology qiang, Shi Jinliang, Zhang Shixu (2009) Spatiotemporal changes
of western and Saharan spring waters of Algeria. Arab J Geosci in soil salinity in a drip-irrigated field. Geoderma 149:243–248
6:665–667. https​://doi.org/10.1007/s1251​7-011-0398-0

13