Evaluation of the quality of groundwater used for irrigation in the Mina plain, Algeria

Gorine, Mohamed; Benkhelifa, Mohammed; Gacem, Farid; Bellague, Djamal

doi:https://doi.org/10.7202/1067306ar

1. Introduction

The irrigated perimeter of Mina extends over 9 592 ha and is located in the north-west of Algeria, about 300 km from Algiers. Agricultural activity is based on the intensive production of cereals and arboriculture. Initially, since 1998, irrigation water (average electrical conductivity : EC = 1.70 dS∙m^-1) came from the Sidi M'Hamed Benaouda dam (SMB), but over time farmers used groundwater more and more (average EC = 5.47 dS∙m^-1) following the drought that has affected the region since 2003.

The poor quality of groundwater compared to that of the SMB dam has led to a process of physical soil degradation (LAHLOU et al., 2002) and a decrease in crop yields by an average of nearly 50% (BENZELLAT, 2012).

The drop in yields in both market gardening (up to 70%) and arboriculture (up to 40%) severely affected the socio-economic situation of the area and the issue of irrigation water quality attracted the attention of farmers and several observers (GACEM, 2014; LAMSAL et al., 1999; PASCAL and BARBAERI, 1995). The challenge is to understand how irrigation water quality affects crop production. For this purpose, it is necessary to establish a diagnosis of groundwater quality. The objective of this article is to provide an in-depth physico-chemical diagnosis of groundwater irrigation quality using two methods, Riverside and Index of irrigation water quality.

2. Materials and methods

2.1 Study area

The Mina plain is located in the north-west of Algeria, at a distance of about 250 km from the capital and 50 km from the sea as the crow flies. It lies between the longitudes 0°17'31" and 0°41'40" E, and the latitudes 35°41'27" and 35°49'07'' N (Figure 1).

Figure 1

**Geographical location of the study area**

**Localisation géographique du site d’étude**

The irrigated plain of the Mina consists of four irrigated areas: Yellel, Mina, Matmar and Oued Djemaa. The plain is characterized by a continental climate with average rainfall of 253 mm∙a^-1 and evapotranspiration of 1 639 mm∙a^-1 (GORINE, 2010).

According to GHOUL and PETER (1974), these soils can be classified into five types: 1) undeveloped soils, 2) halomorphic soils affected by a salt-forming process dating back to the Triassic and Miocene periods, 3) hydromorphic soils affected by physical degradation and no drainage, 4) calcimagnesic soils characterized by high levels of silt, calcium and magnesium and 5) vertisols which contain swelling clay in large proportions. The study area is located in the part of the site that contains calcimagnesic soils and vertisols. The clay content varies between 18% and 49% in the 0-50 cm surface horizon and between 14% and 41% in the lower 50-80 cm horizon. These soils are occupied by cereals, fruit trees, olive trees and market gardening with 46%, 33%, 8% and 4% of the total surface area of the plot respectively (GACEM, 2014).

2.2 Water resources

The Mina plain is crossed in the center by the Mina Wadi and its tributary the Malah Wadi with a very bad quality water, and to the west by the Yellel Wadi. The only source for irrigation of the plain is the water from the Sidi Mhamed Ben Aouda dam, capacity of 225 million cubic metres, for the supply of drinking water and irrigation. Two main aquifers constitute the groundwater resources, namely the Quaternary aquifer and the Astian sandstone aquifer (ABH CZ, 2004).

2.3 Sampling and analyses

The well network, consisting of 178 location-based wells distributed over the entire prospected field, is chosen for the sampling operation in order to acquire representative data on the spatial variability of the area groundwater quality. (Figure 2).

Figure 2

**Well network in the Mina plain**

**Réseau des puits dans la plaine de Mina**

Well water sampling is carried out during the spring-summer irrigation period, from May to September 2008. The water was systematically collected during pumping in 1-L bottles and stored in a refrigerator.

The different analyses on the waters, namely the pH, the electrical conductivity and the ionic balance are carried out at the laboratory of the National Institute of Soils, Irrigation and Drainage (INSID, Matmar).

The chloride, bicarbonate and carbonate ions are made by titration. Sulphate ions are analyzed by the gravimetric method. Sodium, potassium, calcium and magnesium ions are analyzed by the atomic absorption method.

2.4 Assessment of water quality

2.4.1 Riverside method

To assess the quality of groundwater for agricultural use, the Riverside diagram (RICHARDS, 1954) is used to assess the risk of salinization and soil sodization. For this, the sodium adsorption ratio (SAR) parameter is estimated by the following formula:

In addition, the residual sodium carbonate (RSC) parameter also contributes to this assessment. It is estimated after EATON (1950), RICHARDS (1954) and MARLET and JOB (2006) by the formula:

If RSC is positive, it is the alkaline route relative to the precipitation of calcite/sepiolite, otherwise (RSC < 0) it is the neutral saline way. In this case, two cases may occur depending on the signs of the residual alkalinity applied to the precipitation of calcite, sepiolite and gypsum (MARLET and JOB, 2006):

The residual alkalinity becomes positive as a result of the addition of sulphates which will allow the precipitation of gypsum; it is the sulphate-dominated neutral saline pathway;
The residual alkalinity becomes negative even by the addition of sulphates, we speak then of the chloride-dominated neutral saline way.

2.4.2 Irrigation water quality index (IWQI)

Although the water quality index is usually oriented to describe urban water supply, it has been widely used by decision makers in environmental planning (KHALAF and HASSAN, 2013; HAUSSAIN et al., 2014).

It is a simple index that uses the most important parameters that evaluate the quality of irrigation water (YOGENDRA and PUTTAIAH, 2008). HORTON (1965) is the first designer of the water quality degradation indices. Afterwards, several studies were conducted for the measurement of the water quality index. Indeed, ROKBANI et al. (2011), JEROME and PIUS (2010), SIMSEK and GUNDUZ (2007) used the irrigation water quality index (IWQI) as a tool for groundwater quality management.

The IWQI model was developed by MEIRELES et al. (2010) in two stages. As a first step, the relevant parameters for the quality of the irrigation water are identified. In a second step, the definition of the quality values (q_i) and the aggregation weight (w_i) are established. Values of q_i were estimated according to the value of each parameter (Table 1), according to the irrigation water quality parameters proposed by the University of California Consultants Committee (UCCC) and according to the criteria established by AYERS and WESTCOT (1999).

Limit values for quality (qi) measurements (AYERS and WESTCOT, 1999) — Table 1

**Limit values for quality (*q_i*) measurements (AYERS and WESTCOT, 1999)**

**Valeurs limites pour la mesure de la qualité (*q_i*) (AYERS et WESTCOT, 1999)**

Thus the values of q_i are determined by the following formula:

where q_imax : the maximum value of the q_i for the class; x_ij : the parameter observed value; x_inf : the lower limit of the parameter class; q_iamp : the amplitude of the class of the q_i and x_amp : the amplitude of the class of the parameter.

The values of q_i are represented by nondimensional values. The higher is the value, the better is the quality of the water.

In order to evaluate xamp of the last class of each parameter, the upper limit was considered the highest value determined in the physico-chemical analysis of the water samples. Each weight parameter (w_i) used in the IWQI was obtained by MEIRELES et al. (2010) (Table 2).

Table 2

**Weight of irrigation water quality index (IWQI) parameters**

**Poids des paramètres de l’indice de la qualité de l’eau d’irrigation**

The irrigation water quality index is the sum of each water quality value (q_i) multiplied by its corresponding weight and is determined by the formula:

The IWQI, ranging from 0 to 100, is subdivided into classes based on the risk of the salinity problem, the reduction of water infiltration in the soil and the toxicity to the plants (Table 3).

Table 3

**Class characteristics of irrigation water quality index (IWQI)**

**Caractéristiques des classes de l’indice de la qualité de l’eau d’irrigation**

3. Results and discussion

3.1 Descriptive statistics

Physico-chemical analyses of groundwater reveal very large amplitudes between the minimum and maximum values (Table 4).

Table 4

**Physico-chemical characteristics of groundwater in the Mina plain**

**Caractéristiques physicochimiques des eaux souterraines de la plaine de la Mina**

The highest electrical conductivities are close to 16.50 dS∙m^-1, which reflects excessive salinity, due to lithology and high evapotranspiration during the summer period that concentrates the soil solution (CHEVERRY and ROBERT, 1998). Generally, groundwater quality with an average electrical conductivity of 5.47 dS∙m^-1 is considered poor for irrigation. In addition, only nine analyzed wells have an EC ≤ 3 dS∙m^-1 which is the maximum allowable value for most crops (AYERS and WESCOT, 1985). The areas for each class (C4 and C5) are around 50% and have respective values of 44.03% and 55.97% (Figure 3).

Figure 3

**Spatial distribution of groundwater salinity classes (EC: electrical conductivity)**

**Distribution spatiale des classes de salinité de l’eau souterraine (EC : conductivité électrique)**

As for the SAR parameter, there are four classes (Figure 4). Soils assigned to class S1 are located in the north-east and south-west of the study area. On the other hand, class S4 is gathered into a single area located in the eastern part of the prospected area (Figure 5).

Figure 4

**Percentage of soil area for each sodium adsorption ratio (SAR) class**

**Pourcentage des surfaces de sol pour chaque classe de SAR (ratio d’adsorption de sodium)**

Figure 5

**Spatial distribution of groundwater sodium adsorption ratio (SAR) classes**

**Distribution spatiale des classes de SAR (ratio d’adsorption de sodium) de l’eau souterraine**

3.2 Irrigation water quality

3.2.1 Riverside method

The physico-chemical analys of groundwater reveal a salinity unusable for irrigation. According to RICHARDS (1954) classification, modified by DURAND (1983), there are two classes of salinity (C4 and C5) and three classes of sodization (S2, S3 and S4). The Riverside diagram shows a strong dominance of the C4S4 and C5S4 classes with an overall percentage of 88.77%, which is not recommended for irrigation. It is noted that 1.12% of the water is of poor quality (C4S2) and can only be used for well-drained soils and resistant plants with leaching doses. Class C4S3, representing 10.11% of the waters, is of very poor quality and therefore to be used only for exceptional circumstances. That is, only 20 wells are available for irrigation (Figure 6).

Figure 6

**Riverside diagram of groundwater**

**Diagramme de Riverside des eaux souterraines**

3.2.2 Residual alkalinity of water

Without the use of the residual alkalinity indicator, the SAR alone does not show any risk of sodicity for the use, for irrigation, of poor quality groundwater (BOUZADA, 2013), knowing that SAR frequently minimizes the risk of sodization and alkalinization of water in the presence of a chlorinated chemical facies (GOUAIDIA et al., 2012). For this reason, residual alkalinity (RSC) has been chosen as another approach for assessing the quality of these waters. According to the water analyses, there are three residual alkalinity pathways:

RSC > 0, alkaline salinization (RSC1) (1)
RSC < 0, neutral salinization (RSC2) :
- RSC > 0 with sulfated dominance (RSC2.1) (2)
- RSC < 0 predominantly chlorinated (RSC2.2) (3)

In our case, neutral salinization accounts for 83.71% of all water points (Table 5).

Table 5

**Plain water types**

**Types d’eaux de la plaine**

The concentration of bicarbonate ions (HCO₃^-) in water samples is between 0.04 and 14.60 meq∙L^-1 (Table 4). The lower threshold for irrigation water quality standards is 1.50 meq∙L^-1 (AYERS and WESTCOT, 1985). Indeed, an important part of samples records values above the accepted threshold. Under these conditions, the effect of bicarbonates combined with that of salinity favours the infiltration capacity of soils, particularly in the presence of high clay contents (JASSIM and GOFF, 2006).

Chloride concentrations range from 10.00 to 192.95 meq∙L^-1. All samples recorded values above the irrigation water quality threshold of 4.00 meq∙L^-1 (KHALAF and HASSAN, 2013). Under these conditions, chlorides are likely to cause specific toxicity on the crops produced.

The sodium (Na+) ion concentration of the water samples ranges from 8.87 to 100.00 meq∙L^-1 (Table 4). Indeed, 62% of the water samples collected recorded sodium concentration values higher than the threshold allowed (40.00 meq∙L^-1) by irrigation water quality standards (US Laboratory of Riverside). Under these conditions, the risk of physical degradation of soils by sodization must be taken into account.

3.2.3 IWQI method

The result of using the IWQI index, for all well waters in the study area, revealed two main classes: a severe restriction class (SR) and a high restriction class (HR) representing a respective percentage of 64.04% and 36.52%. The index revealed that the Mina plain waters can be used in irrigation with high to severe restrictions.

The method downgraded 39 wells belonging to classes C4S4 and C5S4 to the HR class. Thus, the number of wells available for irrigation (20), classified by the Riverside method, has almost doubled (39).

3.3 Capacity map of irrigation water by the method of intersection between the EC map and the SAR map

The groundwater quality class analysis results show that class (4) C5S (C5S2, C5S3 and C5S4) is the most dominant with a percentage of 73.44% of the total area of 5 678.07 ha. According to the thematic map of water suitability, the C5S class is uniformly distributed in the study area (Figure 7). In addition, the C4S3 (2) and C4S4 (3) are present with percentages of 10.02% and 15.39% (Figure 8), and they are located in the northern and the southern parts of the plain. Class (1) C4S2, limited only to the center, has the lowest percentage (1.15%).

Figure 7

**Thematic map of groundwater suitability of the study area**

**Carte thématique de l’aptitude des eaux souterraines de la zone d’étude**

Figure 8

**Groundwater ability classes**

**Classes d’aptitude des eaux souterraines**

The thematic map established has the advantage of spatially visualizing the ability of groundwater used for irrigation over a large area. It can serve, firstly for the good quantitative and qualitative irrigation management, and, secondly, it is a means of raising users' awareness of what has become of the land irrigated by these types of water.

3.4 Irrigation water suitability map by the IWQI method

Figure 9 shows the spatial distribution of IWQI in the study area and ranges from severe restriction (SR) to high restriction (HR) (Table 3). The field of use of water quality (severe restriction) can be found over a large area of the south-east and east-west of the study area. Severe restriction areas account for 87.55% (approximately 4 971.7 ha). The remaining 12.44% (706.37 ha) is classified as high restriction quality. In addition, the IWQI decreased slightly in the south-east and south-west due to increased electrical conductivity, SAR, sodium and chloride ions. According to the recommendation in table 3, these types of water should be used only with high permeability soil and excessive water application with some constraints on the types of plants for the specified soil tolerance.

Figure 9

**Map of the irrigation water quality index (IWQI) of groundwater in the study area**

**Carte de l’indice de la qualité de l’eau d’irrigation des eaux souterraines de la zone d’étude**

4. Conclusions

At the end of this work, after having carried out a diagnosis of the quality of underground irrigation water in the Mina plain, the conclusions are as follows.

Water salinity is very high compared to the threshold allowed by AYERS and WESTCOT, (1985), EC ≤ 3 dS∙m^-1 for most crops. Indeed, the salinity values obtained show increases ranging from 44.03% to 55.97% compared to the threshold value. These waters, which are commonly used, are likely to affect crop growth and yield.

Depending on the sodization classes, for the values of SAR ranging 10-18, the risk of sodization is moderate and corresponds to 49.5% of the samples studied. Nevertheless, 16.3% of the samples analyzed shows positive residual alkalinity at low risk. The spatial distribution of the EC and SAR, illustrated by the thematic map established by the intersection method, shows that nearly 73.44% of withdrawals concern water of usable quality but with caution.

According to the map established by the irrigation water quality index (IQWI), 87.55% of the withdrawals concern water that can be used for irrigation but in soils with sufficient permeability and by recommending the addition of additional doses of salt leaching at depth. In this case, a functional drainage network is essential, especially as the clay proportions are relatively high.

The determining parameters for irrigation management in the Mina area are the clay content, which controls the water infiltration capacity of the soil and its occupation. The mapping of these two parameters combined with the thematic maps produced as part of this study could provide a valuable database for optimising irrigation and drainage management in order to limit the risks of soil degradation and yield decline. The evolution of saturated hydraulic conductivity in the study site would be very interesting for the control of soil salinity and sodicity at thresholds favourable to crop production.

Evaluation of the quality of groundwater used for irrigation in the Mina plain, AlgeriaÉvaluation de la qualité des eaux souterraines utilisées pour l'irrigation dans la plaine de la Mina, Algérie

Abstract

Résumé

1. Introduction