Towards Sustainable Food Security in the Sahel: Integrating Traditional Conservation Practices and Controlled Irrigation to Overcome Water Scarcity During the Dry Season for Onion and Jute Production

Abstract

In the Sahel, ensuring food security remains a critical challenge due to the region’s prolonged nine-month dry season and the severe scarcity of water resources for irrigation. This study explores an innovative approach integrating two traditional Sahelian soil conservation methods (Zaï and Half-moon) with controlled irrigation to enhance dry-season crop yields, methods not previously explored in combination. A field experiment was performed using a randomized Fisher block design with seven replications assessing the impact of different soil practices on onion and jute production. It also examined the key soil elements and dynamic properties, including N, P, K, pH, temperature, and electrical conductivity. Results showed that the Half-moon technique yields the highest onion production (20.1 t ha⁻¹ ± 0.82), followed by Zaï (18.6 t ha⁻¹ ± 0.48) and flat tillage (14.2 t ha⁻¹ ± 0.84). For jute, the highest third-harvest yield was recorded with Half-moon (9.68 t ha⁻¹ ± 0.63), followed by Zaï (9.56 t ha⁻¹ ± 0.48) and flat tillage (7.53 t ha⁻¹ ± 0.37). These findings offer a viable solution for adapting to climate change by improving water use efficiency and promoting sustainable farming practices in water-limited environments. This research underscores the potential for integrating indigenous knowledge with modern agricultural techniques to mitigate food insecurity in the Sahel.

1. Introduction

In a global context marked by climate change, adaptation measures are becoming increasingly essential for developing countries, where economies rely heavily on agriculture [1]. Faced with the challenges posed by rising temperatures, erratic rainfall, and increasing water scarcity, it is imperative that these countries implement effective strategies to mitigate the negative impacts on agricultural systems and ensure the well-being of their populations. According to the Intergovernmental Panel on Climate Change [2], measures to combat climate change fall into two primary categories: mitigation and adaptation. While mitigation focuses on reducing greenhouse gas emissions, adaptation aims to develop and implement techniques that enable agricultural production under changing climatic conditions. The effectiveness of these strategies depends on several factors, including policy adoption, implementation scale, and stakeholder engagement [3,4].

The impact of climate change on agricultural production is being felt worldwide [4]. Declining crop yields pose severe threats to food security, particularly in regions where populations depend heavily on natural resources and rain-fed agriculture [3]. Predictions suggest that global agricultural production could decrease by up to 30% by 2050 [5]. In Africa, agricultural yields are projected to decline by 15–35% due to rising temperatures, which could increase by 3 to 4 °C [6].

In the Sahel, climate change is exacerbating food insecurity by significantly reducing agricultural production. Research has shown that yields in the region could decrease by 30–50%, depending on location [7]. A study conducted by Sarr et al. [8] highlighted that key Sahelian countries, including Chad, Niger, and Burkina Faso, are experiencing substantial losses in sorghum and millet yields, two staple crops crucial to local food systems. The Sudano-Sahelian region is particularly vulnerable due to uneven rainfall distribution, making it difficult to develop agricultural strategies based on water–soil–plant interactions [9]. Over time, vegetation cover is decreasing due to both climate change and human activities [10], while soil degradation, crusting, and loss of fertility further limit agricultural productivity [11]. Rainfall in the Sahelian zone has become increasingly unpredictable, leading to prolonged droughts that severely impact agricultural development [12,13].

To combat the effects of climate change, many arid and semi-arid regions, particularly in Africa, have adopted soil and water conservation techniques to improve agricultural resilience [14,15]. These techniques include stone rows, grass strips, filter dikes, complementary irrigation, mulching, Half-moon, and Zaï techniques, which vary by country and climatic conditions. These practices loosen crusted soils, enhance hydrodynamic properties, reduce water evaporation, and improve soil fertility [14,15,16,17]. While other techniques contribute to soil conservation, they have limitations. Stone rows reduce runoff and enhance infiltration but require intensive labor and a large quantity of stones. Grass strips help mitigate erosion but have a limited impact on soil moisture retention. Filter dikes improve water infiltration but require frequent maintenance, while mulching minimizes evaporation yet depends on the availability of organic materials. Flat tillage, although effective in loosening the soil, does not promote soil regeneration and instead exacerbates erosion, leading to further land degradation over time. Compared to these alternatives, Zaï and Half-moon techniques reduce evapotranspiration, erosion and soil degradation while optimizing water retention and soil fertility, making them ideal for dry-season farming. Their efficiency in concentrating moisture at the root zone and enhancing plant growth explains their widespread adoption in water-scarce regions [18]. Several studies have demonstrated the effectiveness of these conservation techniques in increasing cereal production. For example, Zaï techniques in Burkina Faso have increased sorghum yields up to 1730 t ha⁻¹ [19], while in Niger, Half-moon techniques have boosted millet yields by 50–100% [20]. Similar success has been reported in Mali, Senegal, and East Africa, where these techniques have significantly improved millet, maize, and cereal production. In Kenya, the use of terraces has doubled maize yields, while in Ethiopia, compost pits have increased cereal yields by 20–30%. In Tanzania and Uganda, Half-moon techniques have led to yield increases of 30–50%. These findings highlight the crucial role of soil and water conservation in strengthening agricultural resilience and food security [21].

Despite the adoption of improved agricultural practices, achieving year-round food security remains challenging due to modernization barriers, regional inequalities, and climate variability [22]. The reliance on rain-fed agriculture, which is limited to a three-month rainy season, exacerbates the issue [23]. While soil conservation techniques have proven effective to combat dry spells during the rainy season, their application to dry-season farming remains largely unexplored. Yet, market crops such as onions and jute play an essential role in local food systems, and their availability is scarce during the dry season [24,25]. Extending their production into the dry season through integrated soil conservation and irrigation is a crucial step toward ensuring food security in semi-arid regions.

This study presents an innovative approach by integrating Zaï and Half-moon techniques with controlled irrigation for dry-season vegetable production (onions and jute) in semi-arid Sahelian conditions, an area of research that has not been previously explored. It evaluates the impact of these techniques on crop yields while analyzing key soil properties, including nutrient availability (N, P, K), pH, temperature, and electrical conductivity. By elucidating soil–plant interactions under varying management practices, this study offers scalable, climate-resilient agricultural solutions, enabling Sahelian farmers to enhance water efficiency, improve soil quality, and strengthen food security in arid environments.

2. Materials and Methods

2.1. Study Area Description

Figure 1 below shows the location of the study site which covers an area of 348 m² and is located in Ouagadougou, Burkina Faso, more precisely on the experimental platform of the agropedology laboratory of the International Institute for Water and Environmental Engineering (2iE) at Kamboinsé. The region’s climate is Sudano-Sahelian, with average annual rainfall varying between 600 and 900 mm. The length of the rainy season is somewhat uncertain, generally beginning in June and extending over a period of 3 to 4 months. The average annual temperature is around 28.7 °C [26,27].

2.2. Description of the Experimental Set-Up

The experimental design adopted is a Fisher block with three treatments and two factors. The treatments include three tillage methods: Zaï, Half-moon, and flat tillage (control). The two factors studied are the crops, onions and jute (Corchorus olitorius). In each plot, we applied the same quantities of water, as well as equivalent quantities of organic and mineral fertilizer. The organic fertilizer was cow manure, while the mineral manure consisted of NPK (15-15-15) and urea. The experimental plots were divided as follows: onions on Zaï, onions on Half-moon, onions on flat tillage (control), jute on Zaï, jute on Half-moon, and jute on flat tillage (control). Each plot measured 6 m² and was replicated 7 times, for a total of 42 plots, i.e., 21 plots for onions and 21 plots for jute. Plots were isolated from each other by polythene film buried at a depth of 50 cm, corresponding to the root depth of the two crops. Plots were also spaced 50 cm from each other to facilitate movement during data collection. The set-up was randomized to minimize experimental errors, such as edge effects, and enable better assessment of the results. Yields were assessed plot by plot, treatment by treatment, and by crop. The set-up was first designed graphically before being implemented in the field. Figure 2 below illustrates the experimental set-up.

2.3. Set-Up of the Cultivation Techniques

This stage preceded the planting of crops and consisted of the installation of the different cropping practices applied in this study, namely Zaï, Half-moon, and flat tillage. All the practices were installed in the plots in January, after the delimitation and isolation of the different plots. The characteristics of the different practices are described below.

2.3.1. The Zaï

The Zaï technique, which has been improved over the years, consists of staggered or straight holes measuring around 20–30 cm in diameter and 10 to 15 cm deep, with spacing that can vary according to the practitioner. For example, Da [19] describes spacing ranging from 50 to 75 cm, while Roose et al. [28] describe spacing of 40 cm. It is therefore a technique that has evolved over time and according to the needs of the practitioner. In this study, the Zaï holes were laid out in straight lines, measuring 20 cm in diameter and 15 cm deep. The holes were laid out in straight, square lines with a spacing of 60 cm between holes and 45 cm between lines. The practices were set up in January. A total of 15 holes were installed per plot, with an exploitable surface area of 3 m² per plot (Figure 3).

2.3.2. The Half-Moon

This is a technique for reclaiming degraded soils that involves digging the soil with picks, pickaxes, and shovels, forming a bowl in the shape of an arc of a circle. The excavated soil is placed on the side of the arc of the semicircle downslope in a semicircular bed with a flattened top [15]. The Half-moons adopted in this study measured 1.5 cm radius, 15 cm depth. The soil excavated from the holes formed a bed approximately 20 cm high. One Half-moon was installed per plot, with an exploitable surface area of 3.53 m² per plot (Figure 3).

2.3.3. Flat Tillage

This is the usual soil cultivation practice adopted when growing onions or jute in Burkina Faso. In this study, it was used as a control plot. The soil was loosened with a pickaxe to a depth of around 15 cm. The exploitable surface corresponded to the surface area of the plot, which was 6 m² (Figure 3).

2.4. Plant Material Description

The crops chosen for this study were onions and jute, both of which have many virtues and are widely consumed locally and around the world [24,29,30,31]. We also chose two crops with different growth and root systems. The onion, which is a bubble plant with a branched or fasciculate root system, and jute, which is a leaf plant with a taproot or spindle root system [32,33], were chosen to appreciate the results of experimentation on plants with different growths. Similarly, in view of the literature and numerous survey results, onions and jute have not been the subject of an experiment combining the working techniques employed in this study. Also, these two crops can be tested together, as they grow on soils with the same characteristics, i.e., not very heavy soils [34,35].

2.4.1. The Onion

Onion (Allium cepa L.), a member of the Alliaceae family, is a biennial plant grown for its bulbs and leaves. Native to Asia, it follows a vegetative cycle of 120 to 160 days [35]. This food is rich in nutrients such as sulfur and the enzymes responsible for its lachrymatory character and pungent taste [36]. In West Africa, onion cultivation plays an essential role in the rural economy, accounting for between 10% and 25% of vegetable consumption [37]. In Burkina Faso, onions rank first in the country’s vegetable production, accounting for over 30% of the total production [38,39,40]. Improved varieties include Texas Yellow Grano, Jaune Hâtif de Valence, Violet de Galmi, Blanc de Tarna, Violet de Soumarana, and a local variety called Violet de Garango [41,42,43]. In this study, “violet de Galmi” (modern variety) produced in Burkina Faso was used, and the seeds were purchased from the market.

2.4.2. The Jute

Jute, also called Indian jute, purple jute, Egyptian spinach, Bulvanka West African sorrel, bush okra, Jewish mallow, tossa jute, délélé, thélélé, or jaillir, is an annual and perennial herbaceous dicotyledonous plant belonging to the Malvaceae family [44,45,46]. It is one of the most popular leafy vegetables in West Africa [24]. It is grown in tropical areas and can be produced over 3 to 4 months, subject to sufficient watering. The sale of jute, whether in powdered or fresh form, is an important source of income [46,47,48,49]. As a vegetable, its root depth should be around 40 cm [50]. In this study, the local (traditional) variety was used, and the seeds were purchased from growers located around Tanghin dam, an area where this crop is particularly intense. Figure 4 shows an image of leaves jute in the farm.

2.5. Plant Set-Up

2.5.1. Setting Up the Onions

The procedure followed in this study draws on various manuals and the results of previous studies, including those of the Ministry of Agriculture of Burkina Faso, the Ministry of Agriculture of Niger (the leading producer in the sub-region), as well as several other documents in the literature review [51,52]. As cultivation took place in the dry season, the local variety Violet de Galmi was chosen in accordance with literature recommendations. The steps are presented in chronological order of their implementation in the field.

The nursery sowing method was preferred over direct sowing and was established on 18 January 2024. The nursery size was determined based on the total field area, with a standard range of 300 to 500 m² per hectare. For the 252 m² experimental site, a 10 m² nursery was allocated. To ensure optimal soil conditions, the nursery bed was disinfected using boiling water (>75 °C) at a rate of 10 L per m², followed by immediate coverage with a polyane film to retain heat. After five days, the polyane film was removed, and 2 m × 1 m seedbeds were prepared, followed by chemical soil treatment with Durexa. Organic amendments in the form of well-decomposed cow manure (1 kg/m²) and NPK (15-15-15) mineral fertilizer (10 g/m²) were incorporated into the soil to enhance fertility. Furrows spaced 20 cm apart were made for planting the seeds, spaced about 1 cm apart, then covered with a thin layer of soil (about 1 cm) to prevent them sinking too deeply. The soil was then lightly watered, leveled, and covered with straw to conserve moisture and regulate temperature. Upon germination (5 days after sowing), the straw was lifted and completely removed once the seedlings attained a height of 5 cm. To mitigate ant infestations, a powdered insecticide (Rambo) was applied around the nursery as a protective measure.

2.5.2. Set-Up of the Jute

After several unsuccessful attempts to launch the jute nursery in January, it was decided to opt for direct sowing in the plots, carried out on the same day as the onion transplanting. The following steps were taken:

• Breaking seed dormancy

As the seeds are dormant, it was necessary to break the dormancy, in accordance with the recommendations in the literature [46]. The seeds were placed in a cotton cloth, then immersed in boiling water (75 °C) for 5 to 10 s. They were then immediately transferred to lukewarm water for 15 min. Finally, they were laid out in the shade for 24 h before sowing.

• Sowing in the plots

The day after dormancy was lifted, seeds were sown in the plots, with 15 cm spacing between seeds and rows. Four seeds were placed in each sowing hole and after about 15 days, once germination was complete, excess plants were removed and only the most vigorous were retained at each sowing point.

2.6. Irrigation Technique Description

The selected irrigation technique is micro-sprinkling, which offers several advantages, including water savings and uniform watering reminiscent of rain. However, it has the disadvantage of being particularly vulnerable to wind [53,54,55,56,57]. As the study site was equipped with a mini meteorological station (Watchdog), the quantity of water supplied corresponded to daily evapotranspiration, multiplied by a factor of 5.26. The average of the daily evapotranspiration during the study was 4 mm. The equation used by Watchdog to estimate the evapotranspiration ET₀ is the Penman–Monteith equation, which is a widely accepted model for evapotranspiration calculations in agricultural studies [58]. The Penman–Monteith equation is the following:

$$E{T}_0={\displaystyle \frac{0.408\Delta \left(R_n-G\right)+y\frac{900}{T+273}{u}_2(e_s-e_a)}{\Delta +y(1+0.34u_2)}}$$

(1)

where ET₀ is the reference evapotranspiration [mm day⁻¹], R_n is the net radiation at the crop surface [MJ m⁻² day⁻¹], G is the soil heat flux density [MJ m⁻² day⁻¹], T is the mean daily air temperature at 2 m height [°C], u₂ is the wind speed at 2 m height [m s⁻¹], e_s is the saturation vapor pressure [kPa], e_a is the actual vapor pressure [kPa], e_s − e_a is the saturation vapor pressure deficit [kPa], and ∆ is the slope vapor pressure curve [kPa °C⁻¹], using the psychrometric constant [kPa °C⁻¹].

In this study, the crop coefficients (Kc) for jute were sourced from existing literature, with values of 0.72 during the initial phase, 1.39 during development, 1.26 at mid-season, and 0.46 at the late stage [1]. To standardize calculations, we used the average Kc (0.96), like the reference evapotranspiration (ET₀) of 4 mm/day, and estimated the net irrigation requirement (Kc × ET₀) over a two-day interval, which resulted in 8.64 mm/2 days. Considering an irrigation efficiency of 70%, the corresponding gross irrigation dose was 12.34 mm/2 days. However, to prevent excessive leaf wilting under intense solar radiation, which could reduce the market value of fresh edible leaves, we increased the irrigation dose to match ET₀ × 5.26, a common practice in the region for market garden crops.

2.7. Maintenance and Harvesting

Before transplanting, the plots received equivalent doses of organic and mineral fertilizers, adjusted to the surface area occupied by the crops. Weeding was carried out regularly according to weed growth and before each amendment. Onion bulbs were harvested 90 days after transplanting, i.e., 125 days after the start of the nursery. The jute plots were treated in a similar way to the onion plots. The amount of basal fertilizer applied per unit area was identical to that used for onions. The main difference lay in the mineral fertilizer applied at each harvest. Harvesting took place every 21 days after the first one, one month after sowing. Each harvest was accompanied by a hoeing operation in the plots.

In the Zaï plots, the surface area really occupied by crops was 3 m², with an application of 2 kg/m² of organic fertilizer (400 g per packet) and 30 g/m² of mineral fertilizer (N, P, K), i.e., 6 g per 20 cm × 20 cm packet. In the Half-moon and flat tillage plots, the crops occupied 3.53 m² and 6 m², respectively. These plots received the same amount of organic and mineral fertilizer as the Zaï plots, i.e., 2 kg/m² of organic fertilizer and 30 g/m² of mineral fertilizer (N, P, K) per unit area occupied by crops. For maintenance fertilization, all the plots received 10 g/m² of N, P, and K 15 days after transplanting, followed by 50 g/m² of urea on days 30 and 45, per unit area occupied by crops.

2.8. Physical and Hydrodynamic Characterization of the Experimental Site

Crop establishment depends on several soil parameters [59]. As the onion and jute crops are dependent on several soil characteristics, a study to characterize the site’s soil preceded the setting up of the experimental set-up and the agronomic phase. On the physical side, parameters such as bulk density, porosity, and granulometry were determined, while on the hydrodynamic side, the hydraulic conductivity at saturation was determined.

The bulk density was determined in 10 cm steps using the cylinder method [60], and calculated by Equation (1):

$$d_a={\displaystyle \frac{m_s}{v}}$$

(2)

where d_a is the bulk density of the soil, m_s is the mass of the dried soil sample, and v is the volume of the cylinder used to take the sample.

Porosity was deduced from bulk density by Equation (2):

$$\%f=(1-{\displaystyle \frac{d_a}{d_r}})\ast 100$$

(3)

where f is the soil porosity, da is the soil bulk density, and d_r is the real soil density; by convention, it is equal to 2.65 g/cm³.

The granulometric study was carried out according to soil horizons in compliance with standard NFP 18-540.

Saturation hydraulic conductivity (Ksat) was determined using the double-ring method [60,61], and the Ksat value was determined using Minitab 17 software.

2.9. Key Mineral Elements and Physical Parameters

The monitoring of key mineral elements and physical parameters in the soil was conducted weekly throughout the plant growth period using the Soil Parameter SpeedMeter, (a portable device enabling rapid in situ measurements). The analysis focused on nitrogen (N), phosphorus (P), and potassium (K) for mineral elements, while the physical parameters included electrical conductivity (EC), temperature (Temp), and hydrogen potential (pH). To ensure representative data collection, five random measurements were taken per experimental plot on each sampling day. The first assessment was performed one week after the transplanting of the onion seedlings and the direct sowing of the jute.

2.10. Data Processing

Site characterization data, i.e., saturation hydraulic conductivity (Ksat), granulometry, bulk density, and porosity, were processed using Excel and Minitab 17.1.0 software.

In this study, bootstrap resampling was applied exclusively to yield data to reinforce the robustness of the statistical analyses, as the original dataset did not fully follow a normal distribution. Specifically, 30 mean values per treatment were generated through resampling before conducting ANOVA and Tukey’s HSD test for multiple comparisons. The Anderson–Darling and Levene’s tests were also performed to assess normality and variance homogeneity, respectively. However, for key mineral elements and physical parameters, principal component analysis (PCA) was directly applied without normality testing or resampling, as PCA is designed to capture variance structures and relationships between variables without requiring data normality assumptions.

To analyze the interaction between the various mineral elements (N, P, K), physical parameters (EC, temperature, and pH), and cropping practices, principal component analysis (PCA) was performed using RStudio software 2024.12.0 version and R software 4.2.2 version. The steps in the procedure were as follows: after standardizing the data, the variables considered included observations of N, P, K, EC, pH, and temperature, while the individuals represented the different cropping techniques. The data were processed separately according to the crops, the soil practice applied, and the measurements taken. In this study, the first and last measurements were studied to better assess the initial and final states of the variables considered [62].

Bootstrap is a statistical method developed by the statistician Efron Bradley, which is a statistical inference technique that allows new samples to be generated solely by drawing and returning data from the original sample [63,64]. Bootstrap data were subjected to several tests, including the normality test of Anderson–Darling [65], the Levene’s test [66], analysis of variance (ANOVA), and finally a Tukey test [67], between the yield values (per unit area of soil occupied by the crops) obtained for different treatments. This was performed to see whether there was a significant difference between the yields of the different treatments, in other words, whether the treatments really influenced the yields or not. The onion yield data relate to weights at the end of the crop. In the case of jute, yields were analyzed separately for each harvesting (depending on whether it was the first, second, or third harvest), followed by an overall analysis of yields for all harvests. The signification level was α = 0.05 for all statistics tests.

3. Results

3.1. Physical Characteristics of the Site Soil

A total of six (06) points were sampled at depths of 10 cm to 50 cm to determine bulk density and porosity. This was performed on a grid basis for a better representation of the results. An average was therefore calculated for all the points according to the different sampling depths. The bulk density (da) showed values ranging from 1.51 g cm⁻³ at the surface to 1.68 g cm⁻³ at 50 cm. Site porosity (f) also varied from 43.02% at the surface to 36.54% at 50 cm. The sampling points (06) for the granulometric analysis of the site soil all showed a sandy-loam texture, with an increase in fine particles with depth. The saturated hydraulic conductivity Ksat was 1.49 ± 0.04 mm/h.

3.2. Yield Evaluation

3.2.1. Onion Yields According to Soil Treatments

From the various statistical analyses carried out, it emerged that soil practices influence onion yields differently under similar dry-season irrigation conditions. With a yield of 20.11 t ha⁻¹, the Half-moon (p-value = 0.60) technique appears to be the best soil practice for onion production in the off-season under micro-sprinkler irrigation. It is followed by the Zaï (p-value = 0.80) technique with an average yield of 18.6 t ha⁻¹. The lowest average yield was observed in flat tillage plots (p-value = 0.39) with 14.2 t ha⁻¹. All the variables followed a normal distribution. Levene’s test (p-value = 0.18) clearly indicates that the variance of the variables is homogeneous. The verification of the two previous tests, coupled with the fact that the average observations in the plots differ from one another, enables us to carry out an analysis of variances. The ANOVA test (p < 0.01) of the onion yields showed just that one of the variable’s variable means is different from the others. To have an appreciation of which variable is different from the others, the Tukey test was performed. The results (p < 0.01) showed that all onion yield averages were significantly different from one another. We can therefore say with certainty that soil treatment influences onion yields differently during the dry season under the same irrigation and fertilization conditions. Figure 5 below shows the box plots of onion yields in different soil practices.

Regarding the interactions between the various components of the soil, the matrix Figure 6a obtained by the principal component analysis (PCA) reveals a high positive correlation between the mineral elements (N, P, K) and the electrical conductivity (EC) of the soil. A moderate positive correlation was also observed between these elements (N, P, K, and EC) and soil pH. There was also a positive correlation between the yields and the variables N, P, K, EC, and pH. On the other hand, a relatively strong negative correlation was observed between mineral elements (N, P, and K) and temperature, as well as between pH and temperature, and between temperature and yields. The combined examination of Figure 6b and Figure 7 show, when superimposed, that for the different soil treatments, the Zaï technique favors a high concentration of the elements N, P, K and a high EC. This is followed by the Half-moon method, while the flat tillage treatment has the lowest concentration of these elements.

Different statistical data for the onion yields are reported in

Table 1. Data from various statistical tests applied to the onion yields.

Variables	Normality Tests		Levene Tests	ANOVA Test	Tukey (HSD) Test
Plots	Tests Applied	(p-Values)	(p-Value)	Pr > F/F	Groups
Onions on Half-moon	Anderson–Darling	0.60	0.18	0.01/527.86	A
Onions on Zaï		0.80				B
Onions on flat tillage		0.39					C

Note: Levene test, ANOVA test, and Tukey (HSD) test at 95% of confidence. For normality tests, the values of p-values greater than the significant level α (α = 0.05) indicate a normal distribution for the variable. For the Levene test, when the p-value is greater than α (α = 0.05), it means that the variables are homogeneous. For ANOVA, when the value of F is greater than the significant level of 0.05 indicated, then the variable means are different from each other. Also, when the value of Pr > F is lower than 0.01, it means that the variables are different. The significant difference is shown by the Tukey test. For the Tukey test, the significant difference is shown by group letter (A, B, C). When two variables present the same letter (i.e., AA, BB, CC), it means that they are not significantly different. Elsewhere, if two variables have different letters, they are significantly different.

. For more details, see the Supplementary Tables S1–S4 for normality, Levene, ANOVA, and Tukey tests.

3.2.2. Jute Yields Depending on Soil Treatments and Harvests

In this study, jute was harvested three times: initially one month after sowing and subsequently every 21 days.

At the first harvest, the highest yield was recorded in the Half-moon plots with 1.96 t ha⁻¹, followed by the Zaï plots at 1.56 t ha⁻¹ and the flat tillage plots at 1.53 t ha⁻¹. Statistical analysis revealed no significant difference between the Zaï and flat tillage yields (p < 0.01). ANOVA indicated that at least one treatment mean differed from the others. The Tukey test confirmed that the yields in the Half-moon plots were significantly different from the yields observed in the Zaï and flat tillage plots which were not significantly different from each other.

At the second harvest, the Half-moon plots again exhibited the highest yield at 8.51 t ha⁻¹, followed by the Zaï plots at 8.01 t ha⁻¹, and the flat tillage plots at 6.53 t ha⁻¹. Normality tests showed p-values > 0.5, indicating a normal distribution while Levene’s test (p-value = 0.32) confirmed the homogeneity of variance. ANOVA results (p < 0.01) indicated significant differences among means, supported by a Fisher value (F = 183.98) exceeding critical values. The Tukey test demonstrated that all treatments were significantly different from one another.

At the third harvest, yields were comparable between the Half-moon and Zaï plots, both significantly surpassing those of flat tillage. The yields recorded were 9.68 t ha⁻¹ for Half-moon, 9.56 t ha⁻¹ for Zaï, and 7.53 t ha⁻¹ for flat tillage. Normality tests (p-values > 0.5) confirmed a normal distribution for the third harvest yields and the Levene test (p-value = 0.33) indicated a homogeneity of variance. These results, along with significant differences in treatment means, justified the application of ANOVA (p-value < 0.01), which indicated that at least one treatment mean differed from the others, as evidenced by a Fisher F value (F = 74.51) above the 0.05 significance level. The Tukey test revealed significant differences specifically between the Half-moon and flat tillage yields, as well as between the Zaï and flat tillage plots. Figure 8 below shows the box plots of the jute yields in different soil practices.

According to the interactions between the various components of the soil, the matrix (Figure 9a) obtained by the principal component analysis (PCA) reveals a high positive correlation between the mineral elements (N, P, K) and the electrical conductivity (EC) of the soil. A moderate positive correlation was also observed between these elements (N, P, K, and EC) and soil pH. There was also a positive correlation between the yields and the variables N, P, K, EC, and pH. On the other hand, a relatively strong negative (higher than for the onion plots) correlation was observed between the mineral elements (N, P, and K) and temperature, as well as between the pH and temperature, and between the temperature and yields. The combined examination of Figure 9b and Figure 10 show, when superimposed, that for the different soil treatments, the Zaï technique favors a high concentration of the elements N, P, K and a high EC. This is followed by the Half-moon method, while the flat tillage treatment has the lowest concentration of these elements.

By overlaying the circle of correlations Figure 9b with the graph of correlations grouping the observations according to the times of measurement and the method of soil treatment Figure 10, several notable variations emerge as a function of these factors. For the flat tillage treatment, a gradual increase in pH and yield was observed, while the average soil temperature fell, after initial measurements showing a soil poor in mineral elements. For the half-moon and Zaï treatments, a high temperature was noted at the first measurement, with a relatively low pH and yield. At the second measurement, the temperature fell, while the pH and yield increased, although the concentration of mineral elements remained low. Finally, at the third measurement, soil properties stabilized, with a particularly high concentration of mineral elements in the Zaï treatment. These results highlight the significant impact of time and methods on soil properties, including nutrient availability (N, P, K), temperature, pH, and yield.

Different statistical data for the onion yields are reported in

Table 2. Data from different statistical tests applied to the jute yields per harvests.

Harvests	Categories	Normality Tests		Levene Tests	ANOVA Test	Tukey (HSD) Test
	Plots	Tests Applied	(p-Values)	(p-Value)	Pr > F/F	Groups
First harvest	Jute on Half-moon	Anderson–Darling	0.29	0.09	<0.01/35.18	A
	Jute on Zaï		0.19				B
	Jute on flat tillage		0.75				B
Second harvest	Jute on Half-moon	Anderson–Darling	0.34	0.32	<0.01/183.98	A
	Jute on Zaï		0.46				B
	Jute on flat tillage		0.72					C
Third harvest	Jute on Half-moon	Anderson–Darling	0.09	0.33	<0.01/74.51	A
	Jute on Zaï		0.54			A
	Jute on flat tillage		0.35					B

Note: This table shows different data from the normality test, Levene test, ANOVA test, and Tukey (HSD) test at 95% of confidence. For normality tests, the values of p-values greater than the significant level α (α = 0.05) indicate a normal distribution for the variable. For the Levene test, when the p-value is greater than α (α = 0.05), it means that the variables are homogeneous. For ANOVA, when the value of F is greater than the significant level of 0.05 indicated, then the variables’ means are different from each other. Also, when the value of Pr > F is lower than 0.01, it means that the variables are different. The significant difference is shown by the Tukey test. For the Tukey test, the significant difference is shown by group letter (A, B, C). When two variables present the same letter (AA, BB, CC), it means that they are not significantly different. Elsewhere, if two variables have different letters, they are significantly different.

. For more details, see the Supplementary Tables S1–S4 for normality, Lavene, ANOVA, and Tukey tests.

4. Discussion

Soil conservation techniques play a crucial role in sustainable agriculture, particularly in semi-arid and Sahelian regions, where climate variability and water scarcity pose significant challenges to crop productivity [14,15]. This study has demonstrated that integrating Zaï and Half-moon techniques with controlled irrigation significantly enhances soil moisture retention, nutrient availability, and, ultimately, crop yields during the growth of onions and jute in the dry season. These findings align with previous studies highlighting the effectiveness of soil and water conservation measures in restoring degraded lands and improving agricultural resilience in Burkina Faso and other semi-arid regions [68,69].

The observed increase in onion and jute yields can be attributed to the ability of Zaï and Half-moon techniques to increase soil infiltration rates, reducing surface runoff and enhancing water and nutrient availability to plants [70]. This synergy optimizes water use efficiency, conserves soil moisture, and improves soil fertility, making these methods valuable in mitigating the impacts of climate change on crop production. Similar studies have confirmed that water and soil conservation practices contribute to improving agricultural productivity by reducing soil erosion, increasing organic matter, and stabilizing soil structure [68,70].

Soil characterization data indicate that the bulk density and porosity values at the study site are favorable for onion and jute production, as these crops thrive in light-textured soils [71]. The correlation between soil mineral elements (N, P, K) and electrical conductivity (EC) aligns with findings from Theresa et al. [72], showing that nutrient dynamics are closely linked to soil electrical conductivity. However, this study found that mineral concentration alone does not directly determine crop yield, as flat tillage plots had lower mineral concentrations but did not record the highest yields. This suggests that other factors, such as soil temperature, soil treatment, and moisture availability, played a role in influencing the results. The higher evapotranspiration rates in the flat tillage plots could have contributed to nutrient losses, as water movement in the soil profile can lead to the leaching of essential minerals. This aligns with the conclusions of Wolka et al. [73] that yield improvements in conservation agriculture result primarily from moisture conservation characteristics rather than from nutrient retention alone.

Historically, prior to the advent of modern irrigation techniques, traditional soil conservation methods such as Zaï and Half-moon were predominantly used during the rainfall season to enhance cereal production and promote forest regeneration in the Sahel, significantly improving soil moisture retention during the dry spell and fertility under harsh conditions [18,74,75]. These soil conservation techniques are also practiced in several countries, primarily in the Sahel region, to enhance cereal yields and restore degraded lands. For example, in Niger, their application has significantly improved millet yields, with increases ranging from 50% to 100% as reported by Sultan et al. [20]. Furthermore, these traditional methods have been implemented in Mali and Senegal, where they have contributed not only to optimizing cereal production but also to promoting land regeneration and forest restoration [18,28]. However, these techniques had not been applied to market garden crop production during the dry season, and by integrating these time-tested practices with controlled irrigation, this study seeks to build on this legacy and offer a modernized approach that enhances crop productivity and sustainability in water-limited regions during the dry season to guarantee year-round food production.

This study revealed that under identical irrigation conditions during the dry season, the Half-moon and Zaï techniques significantly increased onion yields by 5.9 t ha⁻¹ and 3.9 t ha⁻¹, respectively, compared to flat tillage (14.2 t ha⁻¹ ± 0.84). The increase in jute yields across all plots over three harvests further demonstrated the resilience of these techniques in maintaining soil moisture and supporting plant growth over time.

A study conducted in Rwanda by Habineza et al. [71] reported onion yields of 12 t ha⁻¹ using drip irrigation and flat tillage, whereas this study achieved higher yields (20.1 t ha⁻¹ in Half-moon and 18.6 t ha⁻¹ in Zaï plots) despite less favorable soil conditions (higher bulk density and lower hydraulic conductivity). This indicates that water and soil conservation techniques have a substantial impact on soil moisture retention, beyond the irrigation method alone.

The onion yields in this study remained below the global average of 20 t ha⁻¹ but were comparable to national averages, except for the Half-moon plot, which nearly reached the world standard [38,76]. The yields were also higher than those recorded in northern Cameroon (6 t ha⁻¹) in 2003 [77], highlighting the potential for sustainable intensification through conservation techniques.

Despite the promising results observed in this study, further enhancements could be achieved by implementing precision irrigation systems that more accurately meet crop water requirements. Moreover, although the adoption of mechanized Zaï and multifunctional Half-moon techniques could further boost crop productivity, Barro et al. [78] demonstrated that mechanizing Zaï increased seed yields by over 40%, and Nassirou et al. [79] reported that multifunctional Half-moons could triple, or even quadruple, yields compared to traditional methods. These modern approaches demand significantly more application time, involve complex implementation procedures (especially for multifunctional Half-moons), and incur considerably higher costs, particularly in the case of mechanized Zaï.

The large-scale adoption of these technologies is essential for climate change adaptation and food security in the Sahel. Studies have shown that the promotion of soil and water conservation techniques at the community level significantly increases their adoption, leading to improved resilience against droughts and the long-term sustainability of agricultural systems [15,18]. However, several socio-economic factors influence the adoption of conservation techniques, including farm size, household income, access to credit, and knowledge of conservation practices [80]. Addressing these barriers to adoption will be critical to ensuring widespread implementation and long-term sustainability.

5. Practical Applications of the Study

The outcomes of this study offer promising practical applications for enhancing dry-season agriculture in arid and semi-arid regions such as the Sahel. Integrating traditional soil conservation techniques, specifically Zaï and Half-moon, with controlled irrigation can significantly improve water use efficiency and boost crop yields for key vegetables like onions and jute. In practice, this approach enables smallholder farmers to better manage scarce water resources and improve soil fertility, thereby enhancing food security and rural livelihoods. Furthermore, the findings provide a robust foundation for future research aimed at scaling these practices through mechanization and the development of optimized irrigation schedules tailored to local conditions. Future studies should also investigate the economic feasibility and long-term sustainability of these integrated methods, as well as the socio-economic factors influencing their adoption by local communities. This multi-disciplinary strategy has the potential to contribute significantly to climate-resilient agriculture in the Sahel and similar regions facing water scarcity challenges.

6. Conclusions

Ensuring food security in the Sahel requires more than just mitigating climate change impacts; it demands adaptation strategies that optimize water use, enhance soil conservation, and increase agricultural productivity year-round. This study demonstrates that integrating Zaï and Half-moon techniques with controlled irrigation can significantly improve dry-season crop yields, offering a sustainable solution to the challenges of water scarcity and soil degradation. By enhancing moisture retention, nutrient availability, and crop resilience, these techniques provide a scalable approach for semi-arid regions, enabling farmers to cultivate crops beyond the rainy season. Beyond its direct impact on onion and jute production, this approach contributes to long-term environmental sustainability by promoting soil regeneration, reducing land degradation, and improving water efficiency. The findings highlight the potential of traditional knowledge combined with modern irrigation practices to build climate-resilient agriculture in the Sahel. Scaling up these solutions could transform rural livelihoods, stabilize local food markets, and strengthen regional food security in the face of climate variability. Adapting to a changing climate requires innovative, locally adapted strategies, and this research provides a practical pathway toward sustainable agriculture in water-limited environments. By harnessing indigenous techniques and integrating them with efficient water management, Sahelian farmers can overcome seasonal limitations, ensuring year-round food production and building a more resilient future for generations to come. However, the main limitation of this study’s outcomes lies in the challenge of achieving the large-scale adoption of these soil and water conservation practices due to their high labor intensity and socio-economic constraints; therefore, future research should focus on developing mechanization strategies and targeted socio-economic interventions to facilitate broader implementation.