Assessing Yield and Productivity Gaps in Tunisian Maize Cropping System

Abstract

Maize production is deficient in arid countries such as Tunisia. To assess maize yield and estimate productivity gaps among Tunisian farmers in consideration of climate change challenges, a survey was conducted that included 50 farms in 10 governorates, focusing on agronomic practices, seed type adoptions, and socioeconomic parameters. The yield gaps related to water resources and farmers’ technical efficiency represented 26.8% and 32.9%, respectively, while for water productivity, the gaps related to water resources and technical efficiency were 32.2% and 31.3%, respectively. Hybrid varieties were among the 25% yield increase compared to local landraces. Farmers retain local landraces mainly for their food quality. Favorable climatic conditions in the northern regions of Tunisia are among the reasons for higher yield compared to the central and southern areas, which registered a yield reduction of 9.2% and 17%, respectively. The Tobit analyses showed that sowing rate, geographic location, type of variety, and fertilization are the most significant factors contributing to technical inefficiencies. For further increases in maize yield in Tunisia, improving agricultural practices, water management, and using high-yielding varieties are essential.

1. Introduction

Rising temperatures and water scarcity are serious threats to food security in many countries. The C4 crops, known for their high water efficiency and ability to thrive in hot temperatures under current CO₂ concentrations, emerge as a cropping system solution [1]. Among C4 crops, maize (Zea mays L.) stands out as the most widely cultivated cereal, with a production of 1.2 billion T in 2022 and an increased rate faster than that of rice and wheat during the last two decades [2]. However, in North Africa, characterized by an arid and semi-arid climate, maize production represents less than 0.6% of the total world production, compared to 2.1% for wheat [2].

In Tunisia, maize is mainly cultivated as an animal feed source, and smallholders handle most maize areas. Tunisia’s total maize cultivation area was approximately 1240 ha in 2021 [3]. The total maize production remains insufficient to meet the country’s demand, with only 45 T produced in 2022 compared to 827 T imported [2,3]. This production highlights the necessity for enhancing agricultural practices and increased investment to boost maize yields and decrease reliance on imports.

It is worth noticing that local Tunisian maize populations are well-adapted and productive. However, due to several factors, including climate variability, the availability and quality of water resources, and ineffective agricultural practices, reaching optimal maize yields is a challenging objective [4,5,6,7]. The maize yield gaps could be attributed to technological and technical limitations compared to potential levels of production, such as the lack of high-yielding varieties and inefficient farming techniques [8,9,10]. Potential yield (Yp) represents the maximum yield a crop can achieve in a specific environment, determined by the interplay of agroclimatic factors and the crops’ genetic potential, without the constraints of water and nutritional resources [11]. The Yp is typically measured in well-controlled experimental plots or simulated using agroecological models [12]. In contrast, attainable yields (Ya) are achieved by the best-performing farmers, while actual yields (Y) reflect those obtained by farmers using suboptimal agricultural practices [11,12]. Technical efficient yield (Yt) is the maximum yield obtained with limited water resources [9]. Y, Ya, and Yp are indicators used to assess yield gaps [13]. The most commonly used yield gap measurements are related to the ratios of Y/Ya and Ya/Yp [14,15]. The yield gap expressed by Y/Yt, Yt/Ya, and Ya/Yp reflects the technical efficiency, water resource, and technological gaps, respectively [12]. These indicators provide a more meaningful evaluation of yield and crop water productivity than absolute values [16]. Indeed, farmers should aim to minimize the disparity between Ya and Y, known as the intra-gap, which is the key management objective. Narrowing this yield gap requires farmers to assess promising new technologies such as optimizing planting dates [17], addressing pest and disease issues, and nutrient management [8].

In addition to yield gaps, water scarcity is a major constraint to irrigated farming, which consumes over 70% of the water resources in most countries [18]. Thus, water productivity (WP), defined as the ratio between yield and water consumption [19], is used as an indicator of water use efficiency and the farmer’s level of water saving [20]. This indicator is vital for agricultural sustainability in arid areas like Tunisia.

The potential WP values of maize grain are between 4.5 and 5.6 kg/m³ throughout the different continents, while current mean water productivity ranges from 1.1 to 2.6 kg/m³ across these regions [21]. This disparity highlights significant gaps in irrigated water management and underscores the potential for substantial improvements in water use efficiency in maize production.

Maize cultivation in Tunisia has received relatively little attention compared to wheat and barley, and no quantitative assessment of yield gaps, yield levels, and water productivity has been reported in the literature. These milestone gap values are essential to identify the causes of technical inefficiency and make informed decisions on strategies to address the climate change threats [16]. Recognizing the intricate interplay of these elements, our study adopts a comprehensive approach, seeking to unravel the complexities that underlie maize cultivation outcomes in Tunisia by a survey methodology [22,23]. Therefore, the objective of this work was to assess yield and productivity gaps in the Tunisian maize cropping system.

2. Materials and Methods

2.1. Data Collection and Estimations

2.1.1. Data Survey

We carried out a survey involving gathering primary data through interviews, observations, and data collection from a representative sample of maize farms. Key variables such as soil quality, irrigation practices, seed varieties, pest and disease management strategies, and socio-economic factors were evaluated to discern their impact on maize productivity.

The survey was conducted through direct personal interviews with 50 farmers across 10 Tunisian governorates from January to April 2021 and covered 14.3% of the total maize cultivated area, engaging three to seven farmers in each region from the north to the south to encompass different bioclimatic zones (Figure 1). The survey aimed to collect socio-economic data, including each farmers’ age, educational level, cultivated area, and agronomic data, such as yield and technical practices, such as the sowing date, sowing rate, irrigation scheme, and fertigation systems used.

2.1.2. Evapotranspiration Estimations

Mean monthly climate data from 1981 to 2010 were collected from the National Institute of Meteorology’s synoptic stations across all surveyed governorates (https://www.meteo.tn/fr/donnees-climatiques accessed on 18 July 2024). These data included the monthly average maximum and minimum temperatures, average insolation durations, wind speeds, and cumulative precipitations. This information was used to calculate the reference evapotranspiration (ET0) based on the Penman–Monteith formulation [24]. The cultural evapotranspiration (ETc) was estimated by multiplying the crop coefficients (Kc) by ET0 using the Cropwat model:

$$\mathrm{ETc}=\sum \mathrm{K}\mathrm{c}\times \mathrm{E}\mathrm{T}0$$

The maize crop coefficients were 0.3 for the initial stage, 1.2 for the mid-stage, and 0.35 for the last day of the final stage [24]. The maize crop cycle, from sowing to harvest, was reported by farmers to be 125 days as the values reported by Allen et al. [24]. Actual evapotranspiration (ETa) was estimated using the total irrigation amount reported by farmers (I) and the mean cumulative precipitation during the growing cycle (P) for the 1981–2010 period:

$$\mathrm{ETa}=\mathrm{P}+\mathrm{I}$$

Since the maize crop cycle mainly occurs in the summer, the driest time of the year in the Mediterranean area, the cumulative rainfall levels were low and did not exceed 50 mm.

2.1.3. Yield and Yield Gap Estimations

Green fodder biomass yield was estimated at an average dry matter content between 20% and 40%, as the harvest stage varies depending on individual farmer practices, which are influenced by various factors such as climatic conditions and technical approaches. The actual yields (Y) are the yields declared by each farmer in the survey. The maximum green fodder biomass yield, or attainable yield by farmers (Ya), was 70 T/ha. The technically efficient yield (Yt) is the maximum value between the farmers’ declared yield (Y) and the maximum attainable yield when only water is a limiting factor in the production process assessed by the empirical production function [25]:

$$\mathrm{Yt}=\max (\mathrm{Y},{\mathrm{Y}}_{\mathrm{a}}\times (1-{\mathrm{K}}_{\mathrm{y}}\times \left(1-{\displaystyle \frac{\mathrm{E}{\mathrm{T}}_{\mathrm{a}}}{\mathrm{E}{\mathrm{T}}_{\mathrm{c}}}}\right)))$$

where Ky is the yield response factor estimated to be 1.25 according to Allen et al. [24]. The yield gap related to water resources (Gw) is estimated in percentage as:

$$\mathrm{Gw}=(1-{\displaystyle \frac{\mathrm{Y}\mathrm{t}}{\mathrm{Y}\mathrm{a}}})\times 100$$

The yield gap related to farmers’ technical efficiency (Gt) is estimated in percentage by:

$$\mathrm{Gt}=(1-{\displaystyle \frac{\mathrm{Y}}{\mathrm{Y}\mathrm{t}}})\times 100$$

2.1.4. Water Productivity and Water Productivity Gap Estimations

To assess the water use efficiency in maize, various water productivity indexes were estimated. The water productivity is defined according to Molden et al. [26] as follows:

$$\mathrm{W}\mathrm{P}(\mathrm{k}\mathrm{g}{/\mathrm{m}}^3)={\displaystyle \frac{\mathrm{Y}}{\mathrm{E}\mathrm{T}\mathrm{a}}}$$

The attainable productivity $\left(\mathrm{W}\mathrm{P}\mathrm{a}\right)$, the technical efficiency water productivity (WPt), and the farmer water productivity (WP) were estimated using Ya, Yt, and Y as the numerators, respectively [26]. The water productivity gap (G_WPw) related to water resource and technical efficiency (G_WPt) was calculated in percentage as follows:

$${\mathrm{G}}_{\mathrm{W}\mathrm{P}\mathrm{w}}=(1-{\displaystyle \frac{\mathrm{W}\mathrm{P}\mathrm{t}}{\mathrm{W}\mathrm{P}\mathrm{a}}})\times 100$$

$${\mathrm{G}}_{\mathrm{W}\mathrm{P}\mathrm{t}}=(1-{\displaystyle \frac{\mathrm{W}\mathrm{P}}{\mathrm{W}\mathrm{P}\mathrm{t}}})\times 100$$

2.2. Tobit Regression

The Tobit regression analysis was used to investigate the determinants of technical efficiency among Tunisian maize farmers [27,28]. Technical efficiency (TE) scores were estimated using the following ratio:

$$\mathrm{TE}={\displaystyle \frac{\mathrm{Y}}{\mathrm{Y}\mathrm{a}\times (1-\mathrm{K}\mathrm{y}\times (1-\frac{\mathrm{E}{\mathrm{T}}_{\mathrm{a}}}{\mathrm{E}{\mathrm{T}}_{\mathrm{c}}}))}}$$

Since efficiency ranges from 0 to a maximum of 1, the Tobit regression based on the Tobin [28] approach is right censored at 1 when TE exceeds this limit:

$$\mathrm{T}\mathrm{E}=\mathsf{\alpha }+{\sum }_1^7{\mathsf{\alpha }}_{\mathrm{i}}{\mathrm{X}}_{\mathrm{i}}+{\mathsf{\epsilon }}_{\mathrm{i}}$$

where X_i is the independent factor related to the farmer’s technical itinerary, α_i is the regression associated with the X_i factor, and ε_i are independent normally distributed error terms.

The independent factors indexed from 1 to 7 are related to the diammonium phosphate rate (DAP) (kg/ha), ammonium rate (kg/ha), sowing rate (kg/ha), type of seeds (hybrid or local), region (north, center, or south), cultivated area (ha), and the farmer education level (primary, secondary, or high), respectively. DAP, ammonium rate, and the sowing rate are quantitative variables, while the others are categorical variables.

In our study, inferential statistical analysis was performed using the R programming language (R×64 4.1.2), focusing on Tobit regression analysis, also known as a censored regression model, using the package “VGAM” [29]. This approach is used to estimate linear relationships between variables when the dependent variable is subject to left or right censoring.

3. Results

3.1. Descriptive Analysis

The survey data reflect considerable diversity in farming practices and resource use among farmers (

Table 1. Descriptive statistics of the quantitative parameters of the survey made in Tunisia among 50 farmers from 10 governorates.

Variables	Mean	Min	Max
Age (year)	52.6	28	75
Total cropping area (ha)	44.5	2	630
Maize area (ha)	3.6	1	20
Sowing rate (kg/ha)	26.6	10	30
Irrigation (m3/ha)	3660	2500	6000
DAP (kg/ha)	153	100	300
Manure (T/ha)	14	0	40
Ammonium (kg/ha)	200	100	550
Declared average yield of green fodder biomass (T/ha)	34.3	10	70
Water productivity (kg/m3)	8.4	3.5	18.1

). The survey showed that maize farmers had a moderate average age of 52, and 18% were young (age ≤ 40). The total area the respondents manage is variable, even if 74% is less than 20 ha. The area dedicated to maize cropping averages 3.6 ha (Table 1).

The average area for maize cultivation is 3.6 ha, representing only 8.08% of the average cropping area. The sowing rate is relatively consistent, with a mean of 26.6 kg/ha, equivalent to 88.600 grains/ha. Average fertilizer use showed 153 kg/ha of DAP and 200 kg/ha of ammonium. Manure applications depend on farming practices and availability. The same variability was observed for irrigation regimes, with an average of 3660 m³/ha. The results exhibited an average water productivity of 8.4 kg/m³ with a maximum of 18.1 kg/m³ (Table 1).

Similarly, the categorical data showed diverse educational backgrounds, sowing practices, crop rotations, irrigation systems, and final uses among the farmers (

Table 2. Descriptive statistics of the categorical parameters collected from the survey made in Tunisia among 50 farmers from 10 governorates.

Variables	Categories	Farmers (%)
Education	Primary	24
	Secondary	32
	Higher	44
Sowing Date	April	24
	Mai	16
	Jun	24
	July	18
	August	14
	September	4
Previous Crop	Cereals	70
	Legumes	18
	Horticultural crops	12
Irrigation system	Drip	30
	Sprinkler	54
	Submersion	16
Final utilization	Trading	10
	Self-consumption	54
	Self-consumption + trading	36

). The majority of farmers (44%) have attained higher education. This indicates a relatively educated farming population, which may influence the adoption of advanced farming practices and technologies. Sowing dates are spread across several months, with April and June being the most common at 24% each (Table 2). This variation in sowing dates could affect crop growth stages and overall yield. Most early planting dates are concentrated in the southern areas, where temperatures are much higher than in the northern area, reaching 21 °C and 18 °C, respectively, in March. Most farmers primarily choose early planting dates to avoid the period when temperatures peak in the following months, ensuring that maize flowering does not coincide with these peaks. The survey showed that 70% of farmers plant cereals as a previous crop before maize, while crop rotation with legumes and horticultural crops is less prevalent, accounting for only 12% (Table 2). After harvesting cereal crops, the late sowing of maize could allow farmers to gain from a double cropping system.

Regarding the irrigation system, 54% of farmers used sprinkler systems. Drip irrigation systems are used by 30% of farmers, and submersion irrigation methods are used by only 16%. For the final end-product use, more than half of farmers (54%) use their maize production primarily to ensure self-consumption. A significant portion (36%) uses a combination of self-consumption, animal feeding, and selling, while only 10% sell their entire harvest to nearby markets or neighboring farmers (Table 2).

As expected, the hybrid varieties lead to the highest yield with a 25% increase compared to local varieties (

Table 3. Maize yield variability as affected by seed type, region, and irrigation system in 50 farms of 10 governorates of Tunisia.

Variables	Categories	Average Green Fodder Biomass (T/ha)	Standard Deviation (T/ha)
Type of seeds	Hybrid	38.0	11.7
	Local	28.5	8.1
Area	North	36.9	14.3
	Center	33.5	8.7
	South	30.6	5.3
Irrigation System	Drip	39.3	13.1
	Sprinkler	32.4	11.7
	Submersion	31.5	3.5

); otherwise, some farmers still choose the local maize populations characterized by good food quality for their own consumption and/or selling in the local market. Hybrid varieties are often used as forage crops for animal feeding.

In addition, the Northern areas showed 9.44% and 17.14% higher yields compared to the central and southern areas of Tunisia (Table 3). Yield gaps were observed between the three irrigation methods. Drip irrigation exhibited the highest yield, while submersion irrigation registered the lowest yield, showing a difference of 20%.

3.2. Yield Levels and Gaps

The mean calculated ETc was 5100 m³/ha, varying from 3600 m³/ha to 6700 m³/ha depending on the climatic area and sowing date. The results showed that the mean technical efficient yield (Yt) and declared yield (Y) were approximately 51.3 T/ha and 34.4 T/ha, respectively (Figure 2). The levels of yield gap related to water resources ($Gw)$ and to farmers’ technical efficiency (Gt) were 26.8% and 32.9%, respectively. The substantial difference of 35.6 T/ha between the maximum (Ya) and average declared yields (y) indicates a significant overall yield gap of 50%. Therefore, there is a large potential for yield improvement among farmers.

3.3. Water Productivity Levels and Gaps

Relying on the declared yields and irrigation levels, the mean of farmers’ water productivity (WP) was 8.4 kg/m³, and the average attainable productivity (WPa) was 18 kg/m³ (Figure 3). This substantial range indicates an overall gap of 53.5% in water productivity among the farmers (Figure 3). The average technically efficient water productivity (WPt) is about 12.2 kg/m³. The water productivity gap related to water resource (GWPw) and technical efficiency (GWPt) values are 32.2% and 31.3%, respectively. These figures underscore the significant potential for improvement in water use efficiency through better irrigation practices and enhanced water management strategies. By addressing these gaps, farmers could optimize their water usage, leading to more sustainable and productive agricultural operations, particularly within the context of resource shortages.

3.4. Tobit Analysis

The objective of the Tobit regression was to assess the determinants of Tunisian maize farmers’ technical efficiency and to quantify the magnitude and direction of the effects of the factors influencing the technical efficiency of smallholder agriculture.

Tobit analysis revealed that several factors, including fertilization levels, sowing rate, seed type, and region localization, significantly play a decisive role in determining technical efficiency among the surveyed farmers. The sowing rate emerged as the most statistically significant factor (p < 0.01), underscoring its crucial role in the final obtained yield. Additionally, the geographical region factor was statistically significant (p < 0.01); farmers in the southern and central regions demonstrated higher efficiency compared to those in the northern areas where the highest declared yields were registered. This difference could be attributed to the greater availability of water resources in the center and southern regions. Furthermore, the impact of seed variety type was significant (p < 0.05), with hybrids showing an advantage over local varieties (

Table 4. Tobit Regression results for fertilization levels (Diammonium Phosphate and Ammonium), sowing rate, type of variety (Hybrid compared to Local Varieties), region (Center compared to North and South compared to North), cultivated area, and education level among 50 farmers across 10 governorates of Tunisia.

Technical Efficiency	Coefficient	Standard Error	p-Value
Diammonium Phosphate dose (DAP) (kg/ha)	0.000735	0. 000316	0.025 *
Ammonium dose (kg/ha)	0.000541	0.000226	0.021 *
Sowing rate (kg/ha)	0.020889	0.005408	0.000 **
Type of variety (Hybrid/Local)	0.143568	0.057421	0.017 *
Region (Center/North)	0.153858	0.056177	0.009 **
Region (south/North)	0.284737	0.078409	0.001 **
Cultivated Area (ha)	0.005683	0.005090	0.271
Education level (Secondary/Primary)	0.007632	0.064847	0.907
Education level (Bachelor’s/Primary)	−0.074466	0.069895	0.293
_cons	−0.266600	0.159836	0.103
Censoring Parameter (/sigma)	0.143137	0.0151481

Number of observations = 50; Number of censured observations = 4; *: significant at α < 5%, **: significant at α < 1%.

). This highlights the importance of seed selection in enhancing technical efficiency and optimizing agricultural outcomes. Similarly, the amounts of diammonium phosphate and ammonium were significant factors (p < 0.05). This underscores the critical role of basal and nitrogen fertilization in promoting crop growth and productivity. However, no significant effect on technical efficiency is observed for the variables cultivated area and education level (Table 4). Practical farming experience and traditional knowledge might play a more significant role in technical efficiency than formal education levels.

4. Discussion

The findings of this study provide a detailed overview of maize yield gaps in Tunisia, along with factors influencing farmers’ technical efficiency. The analyzed quantitative and categorical variables reveal significant trends in maize agricultural practices and farmers’ performances.

The maize date of sowing is highly variable and can critically impact yield. Nearly a quarter of the surveyed farmers prefer early sowing and choose April to avoid the heat waves of July with a national average temperature of 33 °C, which could negatively affect flowering, thus reducing the final number of kernels [30], because high temperatures around the silking period result in significant reductions in grain yield in summer maize. Early sowing in the spring is mainly adopted in the southern regions to avoid severe heat waves where temperatures could reach more than 45 °C in July. At the same time, some farmers started with early sowing and, upon completion, proceeded with late sowing, taking advantage of maize’s short growth cycle, which ranges from 95 to 120 days. In that case, the farmers can profit from double cropping of the field by doing the early sowing in April and late sowing in September. Maize can be sown twice per season, depending on the agronomic aptitude of the variety, the end-use intended product, and soil moisture and weather conditions. In the northern regions, late sowing is also preferred due to better rainfall, which enables farmers to benefit from the autumn rains and reduce irrigation costs. The importance of sowing dates has been demonstrated by earlier research [31]. Otherwise, this study did not register a significant effect of the sowing date on the technical efficiency. This result could be attributed to climatic differences between the cooler northern regions and the hotter southern regions.

Besides climate, water availability is a critical factor that varies between governorates, which impacts yield with a maximum in the northern regions. Tobit’s analysis revealed that farmers in the southern and central areas are more technically efficient than those in northern Tunisia. Farmers in the south and central regions benefit from private irrigation water resources, relying on groundwater and surface or deep wells, which are more reliable resources than surface water allocated through the network of agricultural development groups responsible for the distribution of surface water in Tunisia [32]. In contrast, most northern areas depend on public irrigation water sourced from reservoirs fed by rainfall [33]. Recent climatic conditions and disrupted rainfall patterns have made water availability less predictable, leading to irrigation disruptions. These findings are consistent with those reported by Chemak et al. [33], highlighting the importance of free access to resources compared to collective management. However, free access to water resources can lead to overconsumption and unwise use, negatively impacting the water productivity indicator [34].

The adopted irrigation system is a key determinant element for maize yields. Drip irrigation is not the most used method for maize, while most farmers opted for sprinkler irrigation, possibly due to its lower cost. However, some farmers still use submersion irrigation even if it is an outdated method that leads to uneven water distribution and unnecessary wastage, promoting weed growth and fungal diseases [35]. Economic considerations must also be considered, as adopting drip irrigation may entail additional costs for farmers. Previous studies [35] indicate that, despite its negative environmental impacts, submersion irrigation is preferred for its low operational costs. Nonetheless, this reliance on submerging and sprinkler irrigation may contribute to the observed water productivity gaps [36].

Crop rotation practices can also influence the farming technical efficiency. Thus, crop rotations incorporating legumes and horticultural crops are relatively efficient for improving yield and efficiency; 70% of the farmers planted maize after cereals. Compared to rotating with horticultural crops, cereal-to-cereal rotation can degrade soil fertility and increase the risk of weed proliferation and disease. Previous research has produced similar results, supporting this perspective, emphasizing the determinant impact of inappropriate rotation strategies [37]. Otherwise, according to Tobit analysis, the direct effect of rotation on yield is not evident, possibly because it could be masked by other factors such as fertilization and sowing rates.

The results underscore the importance of increasing the sowing rate to improve the final yield. Increasing the sowing rate can improve the maize’s total biomass and yield per unit area. However, escalating plant density can intensify intraspecific competition and maize plants’ growth and development [38]. The survey findings highlighted the benefits of promoting more productive hybrid seeds instead of local populations [39].

Diammonium phosphate (DAP) and ammonium-nitrate (NH₄⁺-NO₃⁻) are essential fertilizers for improving maize technical efficiency, supporting optimal nutrient uptake. Assefa et al. [40] reaffirmed the positive impact of balanced fertilization on root and shoot plant growth, ultimately boosting yield. In fact, the role of providing adequate fertilization of nitrogen (N) and phosphorus (P) in enhancing early plant establishment and biomass production and their crucial effect for achieving high productivity was reaffirmed by several researches. In fact, applying DAP at planting represents an available source of phosphorus to stimulate root growth, while split applications of ammonium nitrate during vegetative and reproductive maize growth stages supply nitrogen when maize demands are highest [41]. Combining DAP and ammonium on a balanced fertilization can significantly optimize yield potential under diverse conditions [42].

The results also revealed that 56% of surveyed farmers have a lower educational level than a bachelor’s degree. Still, no significant effect on the Tobit analysis was registered for this factor in respect to the technical efficiency gap, indicating that the farmers could have good agronomic expertise regardless of their educational background. The findings contradict the conclusions drawn by other results [43,44], who highlighted that education positively influences technical efficiency. Otherwise, education could help to improve the current situation because farmers with a high educational level are more receptive to communication and willing to adopt new technologies, as emphasized by many authors [45].

The Tobit analysis results for cultivated areas did not register any significant impact on technical efficiency, showing that the agronomic field management to achieve technical efficiency is unrelated to the cultivated area. This result may be found because most of the land dedicated to growing maize is relatively small. This result was in contradiction with different publications for other crops [40], who evaluated the technical efficiency of Durum wheat in Tunisia and found that the technical efficiency was related to small farm size, which should be easily managed. The analysis of yield and water productivity gaps has underscored that both water resources and technical efficiency contribute significantly to the total yield gap. This observation aligns with findings from previous studies documenting crop yield gaps [10,17]. The disparity in water productivity levels indicates inefficiencies in water use across different farming practices. Farmers achieving higher water productivity are likely employing better irrigation techniques, crop selection, and water management strategies.

The large yield gap shows that many farmers are not achieving their full yield potential due to various factors such as suboptimal management practices, inadequate inputs, or environmental constraints. The recorded overall yield gaps are comparable to those observed in other developing countries with climatic conditions more suitable to maize cultivation. For example, in northern Ghana, which has a tropical climate with an annual rainfall average of 900 mm and temperatures ranging from 25 °C to 30 °C, the yield gap for average farmers ranges from 59% to 75% compared to the optimal yield, while water-limited yield gaps range from 67% to 84% [46,47,48]. Similarly, in central Malawi, with a subtropical climate, technical yield gaps are around 36.9% [45]. In five districts of Ethiopia, where temperatures range from 15 °C to 30 °C and rainfall varies from 800 mm to 2200 mm, water-limited yield gaps range from 15% to 73% depending on the farming system [42].

A comparison of water productivity (WP) between Tunisia and other Mediterranean regions revealed that the average WP for rainfed maize in these regions was about 42% below the potential WP [44]. Typically, maize WP gaps range from 20% to 46% of potential values. Zheng et al. [21] reported that the global mean WP for maize production was 18.6 kg/m³, with values varying from 8.0 to 33.2 kg/m³. WP was found to be higher in Europe compared to other regions. In Africa, where seasonal water supply is lower than in the reported European regions, maize yields are also lower, resulting in reduced WP. This indicates significant untapped potential to enhance water productivity and yields in Africa through improved management practices [21].

Addressing these gaps through improved agronomic practices, better seed varieties, and enhanced management strategies could lead to substantial increases in agricultural productivity and food security.

5. Conclusions

Technical efficiency is a crucial factor leading to an increase in yield and farmers’ incomes, along with ensuring sustainable resource use and thus food security. The results reported that sowing rate with an average of 88.600 grains/ha emerged as the most significant factor on technical efficiency, emphasizing the need to improve plant density. Although even the highest yields were registered in the northern regions, the southern and central regions demonstrated higher technical efficiency. Hybrid varieties recorded higher yields compared to local varieties, which are still used mainly for their food quality. Basal and nitrogen fertilization using DAP and ammonium play an important role in promoting technical efficiency. The results reported that sowing rate, geographic location, type of variety, and fertilization are the most influential factors of technical inefficiencies. These elements represent the most critical aspects that decision-makers should prioritize to enhance maize farmer performance. The variability in water availability due to climatic conditions further underscores the importance of improving irrigation resources for maintaining and improving agricultural productivity. This study highlights the need for region-specific solutions to farmers’ unique maize cropping challenges.