Effects of Inequality of Access to Irrigation and Water Productivity on Paddy Yield in Nigeria

Abstract

This study assessed the equity in irrigation water uses and its impacts on paddy yield and water productivity among rice farmers in the Kano River Irrigation Project. Two hundred and twenty-five (225) respondents were used for the data collection from January to June 2021. The data were analyzed using the Kruskal Walli’s test, Tukey kramer post-Hoc test, physical water productivity, and the Logit model. The results indicated that the downstream farmers had the lowest mean yield (1625 kg/acre), lowest access to irrigation water, and lowest irrigation water use (2430 m³/acre). However, they had the highest water productivity (0.66 kg/m³) against 0.44 kg/m³ and 0.58 kg/m³ for middle and upstream farmers, respectively. The logit regression results disclosed that the farm locations, quantity of irrigation water, access to irrigation water, and training on water use were statistically significant determinant of paddy output. Efforts to teach farmers about the actual crop water requirements are needed to promote irrigation water efficiency and conserve scarce resources for other competing users.

1. Introduction

Despite the contributions of irrigation to poverty reduction and food security, irrigated agriculture contributes less than 12% of the total cropped area in Nigeria. Low crop water productivity in upstream farms and high crop water productivity in downstream farms suggest the need for improved technical efficiency, especially among the farms located upstream. Disparities in access to irrigation are serious challenges for the sustainability of the large-scale irrigation project in Nigeria. Bridging thus gap in equality can increase rice production substantially, thereby reducing poverty and food insecurity and increasing income.

Irrigation plays a critical role in the livelihood of farming communities all over the world [1]. Many works have indicated that distributional heterogeneity prevents large-scale irrigation projects from reaching their full potential [2]. Disparities in farmers’ abilities to benefit from irrigation resources are likely to cause distributional conflicts and factionalism [3]. Insufficient irrigation water supply (IWS) downstream mostly leads to lower crop yields compared to the farms located upstream. Downstream farmers in some irrigation systems cannot plant water-intensive crops such as rice; therefore, they instead cultivate crops with lower water requirements like corn and soybeans [4]. Numerous studies have indicated that downstream farmers face more uncertainty and risk because they receive less water than farmers located upstream [5,6]. Therefore, some studies [7,8,9,10] have opined that unless water quotas are carefully enforced, upstream farmers will continue to consume more water at the expense of downstream farms.

Access to irrigation water and water productivity has a cause/effect relationship [11]. Excess irrigation can result in less water productivity due to over-irrigation and unproductive water outflows, seepage, percolation, and evaporation. Acute water shortages can also lead to less water productivity due to moisture stress, while moderate water use can enhance water productivity and water use efficiency. Therefore, access to irrigation water is a necessary but not sufficient condition for higher crop water productivity.

Other factors that enhance crop water productivity include good agronomical practices and judicious use of other inputs, including the use of improved cultivars and the efficient use of fertilizers. Increased output may arise from improved yields, reduced crop loss, improved cropping intensity, and increased cultivated areas. Reliable access to water enhances the use of complementary inputs, such as high-yielding varieties and agrochemicals, which also increase output levels and improve crop water productivity [12]. In many large-scale irrigation projects, water is available but does not produce potential gains due to lower water productivity. Resource mismanagement, especially by upstream farmers, and acute inadequate water downstream are the main causes of lower productivity. Reliable access to irrigation water not only raises output levels and water productivity but also reduces variance in output across seasons [13].

A plethora of research on irrigation management has claimed that upstream farmers tend to over-irrigate their fields without considering their downstream counterparts [6,14,15]. According to [14], downstream farmers are enraged by upstream farmers who use water directly from the distributary and drainage canals to cultivate their paddy fields. The most troublesome issues for downstream water users are illegal water drawing and the consequent farmer discord [16]. A Study by [10] postulated that water distribution disparity, in addition to transportation losses and technological constraints, leads to farmer conflicts. However, inefficient management and unequal distribution of water have led many experts to conclude that large irrigation projects, especially in the developing countries like Nigeria, fall short of achieving their targeted objectives [17].

The linkage between irrigation, poverty, and food security in developing countries had been widely discussed by researchers and policy makers [18,19,20,21]. However, competition for limited water resources and disagreements over water distribution among diverse water users are significant challenges to the sustainability of large-scale irrigation system management across developing countries [22]. Although Nigeria has made considerable progress in irrigation water resources and development towards improve food security, income, and economic growth [23], empirical findings indicate that large-scale irrigation projects in the country have favored some farmers, particularly those upstream, expanding wealth disparity [1].

The Hadejia-Jama’are River Basin Development Authority established the Kano River Irrigation Project (KRIP) in 1976. The KRIP and other large-scale irrigation schemes have been mandated to increase the country’s aggregate food and raw material production while increasing farmers’ incomes, livelihoods, and food security [24]. Increased revenue will improve the quality of life of farmers and others engaged in the value chain generated by irrigation. Reference [25] claimed that examining the internal and external aspects affecting water supply performance downstream is vital to understanding each scheme’s unique potential. The KRIP has 18 sectors, each with a main canal (MC), branch canals (BCs) and distributary canals (DCs). The project was designed to distribute water equitably to all farms in the project area via gravity. However, due to managerial problems, siltation, and farmers’ attitudes towards breaking the field canal, the downstream farmers receives less water, making them prone to crop failure and receiving less income than their counterparts. Furthermore, In the 1970s and 1980s when Nigeria was seeing substantial income from oil exports, the government was able to offer cash for maintenance and operation (M&O) of the irrigation project. Therefore, the farming community handed off responsibility for this task to the government. However, due to competing demands from other sectors of the economy and the diminishing oil income, these subventions gradually dwindled each year. Thus, the water distribution infrastructure deteriorated as it went without maintenance for too long. This led to a decline in the scheme’s effectiveness, eventually affecting the farmers’ livelihoods. Reference [26] showcased that there is a positive relationship between water inequality and income inequality and between water inequality and rural poverty.

References [6,27,28] have shown that water availability, dependability, efficiency, and equity negatively affect the performance of downstream farmers. However, few studies have proven how water access affects productivity and income inequality [4,29]. Similarly, numerous studies have also used various indicators to measure water delivery performance [30,31,32,33,34,35,36]. References [37,38] employed quantitative analytical methodologies to evaluate the access to irrigation water in large-scale public irrigation.

Moreover, a comprehensive study of access to irrigation water in large-scale public irrigation among the diverse water users in Nigeria is not yet available. This article intends to: (a) assess upstream and downstream equity in water distribution and its impact on paddy output and water productivity in the Kano River Irrigation Project, Kano State, Nigeria, (b) provide empirical evidence of the degree of water productivity between the upper- and downstream farmers, (c) present some promising pathways for improving water distribution and equity for poverty reduction among irrigated farmers, and (d) suggest policy measures that will provide solutions that help to close the productivity gap, expand irrigation water availability, reduce resource mismanagement, and boost agricultural production across the board. In light of these concerns and pursuits, this study aims to test the following hypotheses:

H1. 

The distribution of irrigation water use is the same across location categories.

H2. 

The distribution of access to irrigation water is the same across farm location categories.

H3. 

The output produced is the same across farm location categories.

Conceptual Framework

To eradicate poverty, improve food security, and spur economic growth among irrigated farmers, farmers should have equal access to irrigation water and the benefits that come with it [26]. Recently, in Sub-Saharan Africa, there has been a resurgence of interest in the subject of how irrigation expansion might help reduce poverty [39]. However, more robust policy interventions are needed to reduce the disparity in the distribution of social and economic gains among farmers in irrigation areas, as wealth creation alone does not always alleviate poverty. These persisting inequalities may limit the possibilities for additional poverty reduction through economic growth and compromise the impact of governmental measures [40].

Reference [41] opined that the availability and accessibility of irrigation water is a necessity, but not a guarantee, for alleviating poverty. They further claimed that the final result depends on the relationships between water and non-water sector concerns and processes, which can either alleviate or exacerbate poverty. Reference [42] outline five broad factors that influence the impact of poverty eradication on irrigated agriculture. These include: (a) equitable land acquisition; (b) irrigation infrastructural management; (c) irrigation water availability and equity in access to irrigation water; (d) production technology, cropping patterns, and crop diversification; and (e) input- and output-based support measures. Figure 1 presents a nexus between poverty traps and inequality in access to irrigation water among irrigated farmers.

Irrigation water disparity can lead to low productivity and low income, leading to low levels of demand and savings (as shown in Figure 1). Low savings and low market demand can lead to low physical and human capital investments, resulting in increases in poverty levels and long-term food insecurity.

Reference [11] claimed that the incidence and severity of poverty among resource-poor irrigated farmers depend on their level of control over water resources rather than on their resource endowment. They cited an example from Eastern India, which is endowed with a very large groundwater reservoir and substantial surface water resources; however, people lack the resources to exploit these water sources, resulting in poverty in the upstream regions of India. Reference [30] demonstrated that, although the underlying causes of poverty vary based on farming systems, the increasing scarcity and competition for water pose a threat to future advances in poverty reduction in many countries, especially in the South Asia and SSA. Indeed, most of the areas of persistent poverty can be described as “water scarce”. However, many irrigated areas with large-scale systems remain home to large numbers of poor people in both absolute and relative terms. This is largely due to inequity in access to land and water resources, resulting in low productivity, particularly in downstream areas [43]. Reliable access to agricultural water not only raises crop output levels but also usually reduces variance in output and income over time. For instance, in Brazil, the entropy index of rice yield dispersion, which is a measure of yield variability, has declined from 5.3 in 1975 to 2.7 in 1995 in irrigated areas, while in rain-fed areas, it has increased from 8.0 to 13.7 over the same period. Lower entropy index values indicate a lower yield variability, while higher values indicate a higher yield variability. Moreover, the mean yield difference between irrigated and rain-fed areas has also widened [44].

2. Materials and Methods

2.1. Study Area

Three-quarters of Nigerians (out of 203 million) rely on agriculture directly or indirectly for their livelihoods, which accounts for 23% of the country’s GDP [45]. Even though Nigeria is the largest economy in Africa, approximately 79 million (39.1%) Nigerians lived below the poverty line in 2019. Of these, 84.6% were living in rural areas and 76.3% were living in northern Nigeria [46]. This article is based on data collected from one of the largest public irrigation schemes, the Kano River Irrigation Project, located in the north-western region of the country (Figure 2).

The KRIP is located 08°45 degrees east of Greenwich and 11°45 degrees north of the equator. It is one of the country’s functional irrigation systems, and it is located in a massive region more than 25 km south of Kano city. With the help of water from the Tiga dam, the project is able to irrigate close to 22,000 acres of land. The project covers three local government areas: Kura, Garum Mallam, and Bunkure. Garum mallam is located upstream, while Kura and Bunkure are located in the middle and downstream, respectively. All three LGAs were included in this study. From the project design, “it was acclaimed that the whole project is to be operated by gravity. Farmers in the command area were expected to use siphon only for water application to the farmlands, and the siphon diameter should be 65 mm”.

Based on lengths of the main, branch, and distributary canals, they were sort into three groups: upstream, middle, and downstream (

Table 1. Categorization of the Farms in the Study Area.

Type of Canal and Total Length	Upstream Area	Middle Area	Downstream Area
Main canal (25 km)	First 8.3 km	Between 8.4 and 16.6 km	Between 16.6 km and 25 km
Branch canal (27 km)	First 9 km	Between 9 and 18 km	Between 18.1 and 27 km
Distributary canal (204 km)	First 68 km from the branch canal	Between 68 and 136 km from the branch canal	Between 136 and 204 km from the branch canal

).

2.2. Sampling Procedure

The data used in this study were acquired between January and June 2021 from 225 farming household interviewed by three local governments (Kura, Bunkure, and Garum Mallam) in the project area. The farmers were classified into three groups: irrigators at the upstream, middle, and downstream areas. Through stratified sampling, seventy-five (75) farmers were chosen from each location; thus, 225 rice farmers formed the sample size. However, 207 valid responses were retrieved and used for the analysis. A questionnaire and physical water measurements were used as the instruments for data collection. A representative member of each household (typically upstream) was questioned following a predetermined list of questions regarding water supply, economic disparity, agricultural productivity, and paddy output. In relation to the yield of the paddy, the respondents were asked about the number of bags they produced per acre, which was converted into kgs using a standard of 75 kg per bag.

Qualitative (descriptive) inquiries were used to elicit information about the farmers’ views and perceptions. Close-ended questions, such as Likert-type rating scales, and open-ended questions eliciting narrative responses yielded themes that were then synthesized.

Measurement of Irrigation Water Use

Following [47], the bucket method was adopted to estimate the quantity of water used by the sampled farmers during the 2021 dry cropping seasons. This method used a stopwatch to collect the water flow through the siphons in a plastic bucket of 22 L. The bucket was implanted into a small excavated pit with a surface approximately equal to the surface of the irrigated plots to attain consistency.

The following formula was used to obtain the total water application for the cropping season:

$${\mathrm{X}}_{\mathrm{w}}={\mathrm{G}}_{\mathrm{r}}.{\mathrm{D}}_{\mathrm{h}}.{\mathrm{F}}_{\mathrm{w}}$$

(1)

where:

${\mathrm{X}}_{\mathrm{w}}=\mathrm{A}\mathrm{p}\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{x}\mathrm{i}\mathrm{m}\mathrm{a}\mathrm{t}\mathrm{e} \mathrm{a}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t} \mathrm{o}\mathrm{f} \mathrm{w}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{r} \mathrm{a}\mathrm{p}\mathrm{p}\mathrm{l}\mathrm{i}\mathrm{e}\mathrm{d} \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{c}\mathrm{r}\mathrm{o}\mathrm{p}\mathrm{p}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{s}\mathrm{e}\mathrm{a}\mathrm{s}\mathrm{o}\mathrm{n} \mathrm{i}\mathrm{n} {\mathrm{M}}^3$;

${\mathrm{G}}_{\mathrm{r}}=\mathrm{D}\mathrm{i}\mathrm{s}\mathrm{c}\mathrm{h}\mathrm{a}\mathrm{r}\mathrm{g}\mathrm{e} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e} \mathrm{i}\mathrm{n} \mathrm{l}\mathrm{i}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{s} \mathrm{p}\mathrm{e}\mathrm{r} \mathrm{s}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{d}$;

${\mathrm{D}}_{\mathrm{h}}=\mathrm{F}\mathrm{a}\mathrm{r}\mathrm{m}\mathrm{e}\mathrm{r}\mathrm{’}\mathrm{s} \mathrm{d}\mathrm{a}\mathrm{i}\mathrm{l}\mathrm{y} \mathrm{h}\mathrm{o}\mathrm{u}\mathrm{r}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{w}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{r} \mathrm{a}\mathrm{p}\mathrm{p}\mathrm{l}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}$ (the time taken to irrigate farm land in hours);

${\mathrm{F}}_{\mathrm{w}}=\mathrm{F}\mathrm{r}\mathrm{e}\mathrm{q}\mathrm{u}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{y} \mathrm{o}\mathrm{f} \mathrm{w}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{r} \mathrm{a}\mathrm{p}\mathrm{p}\mathrm{l}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}$ (How often the farmer irrigates his/her farm land per week).

The actual amount of water applied per hectare was:

$$\frac{{\mathrm{G}}_{\mathrm{r}}.{\mathrm{D}}_{\mathrm{h}}.{\mathrm{F}}_{\mathrm{w}}}{\mathrm{s}\mathrm{i}\mathrm{z}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{f}\mathrm{a}\mathrm{r}\mathrm{m} \mathrm{i}\mathrm{n} \mathrm{h}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{s}}$$

(2)

The discharge rate was estimated by applying Michael’s (1981) formula:

$${\mathrm{G}}_{\mathrm{r}}=\raisebox{1ex}{$\mathrm{V}$}\!\left/ \!\raisebox{-1ex}{$\mathrm{T}$}\right.$$

(3)

where:

G_r = Discharge rate litres per second;

V = Volume of the container in liters;

T = Time taken in seconds to fill the container.

The discharge rate was multiplied by the farmers’ number of hours of water application per day. The observed daily quantity was multiplied by the water application frequency for the water application period to obtain the farmers’ estimated total amount of water. The total volume of water used was estimated in cubic meters and then divided by the farm size (in acres) to obtain the actual volume of water applied per acre. The daily hours of water application (Hh) is the number of hours a farmer spent irrigating an acre of farm land per day. The frequency of water application is the number of times a farmer irrigates his or her farm per week.

The technical water productivity (kg/m³) is calculated by dividing the seed yield (kg/ha) by consumptive usage (m³/ha):

$$\mathrm{PWP}={fx}^{-1}$$

(4)

where f is the output produced in kg and $x$ is the quantity of water applied in m³.

The ratio of the gross value of the product to the volume of irrigation water used is known as the economic water productivity.

2.3. Data Analysis

2.3.1. Kruskal–Wallis Test

The Kruskal–Wallis test was used for with one-way data containing more than two groups. In the absence of additional assumptions about the data distribution, the test does not address hypotheses about the group medians. It is used to determine whether an observation made in one group has a higher likelihood of being significant than an observation made in the opposite group. This can be determined if one sample has stochastic dominance over the other. The test assumes that each observation is independent of the others. This method cannot use observations with paired or repeated measurements [48]:

$$H=\frac{12}{n\left(n-1\right)}\sum \frac{{Ri}^2}{ni}-3\left(n+1\right)$$

(5)

where:

H is the Kruskal–Wallis test statistics;

n is the total number of variables;

Ri² is the rank total for each group (irrigation water use, access to irrigation water, paddy output);

ni is the number of variables in each group.

2.3.2. Post Hoc Tukey–Kramer Test

The Kruskal–Wallis test does not specify which variables are significantly different from one another. A post-hoc comparison using Turkey’s HSD was conducted to determine the pairwise significance as follows:

$$HSD=qA,\mathsf{\alpha },dof\sqrt{\frac{{MS}_E}{2}}\left(\frac{1}{ni}+\frac{1}{nj}\right)$$

(6)

where:

qA,α,dof = student size range statistics with A number of groups and α significance level;

MS_E = square mean error;

n = groups sample size (i and j).

2.3.3. Logistic Regression Analysis

This research used logistic regression analysis to predict the probability of paddy output obtained by the farmers in the KRIP. The calculation assumes that the paddy output obtained is a function of socio-economic and institutional factors. This binary logistic regression methodology has been employed in several agricultural, economic, and extension studies that call for the research and prediction of a dichotomous outcome, such as fertilizer use or non-use, adoption and non-adoption, and participant and non-participant [49,50,51]. The detail of explanatory variables is presented in

Table 2. Definitions of explanatory variables.

Variable	Explanation	Expected Sign of Variable
LCTN	Dummy variable of location (1 = upstream, 0 = otherwise)	+ or −
WUA	Dummy variable of functional water users association (1 = yes, 0 = no)	+
IT	Dummy variable of illegal tapping (1 = yes, 0 = no)	−
CRC	Condition of the distr. canal (1 = yes, 0 = otherwise)	?
EDU	Level of education of the farmers (continuous variable)	+ or −
GEND	Dummy of gender (1 = male, 0 = female)	+ or −
TWU	Dummy variable on training on water use (1 = yes, 0 = no)	+
AIW	Access to irrigation water (1 = high, 0 = low)	+
IWU	Quantity of irrigation water use in M3 (continuous variable)	+
FEXP	Farmer’s experience (continuous variable)	+ or −
RECP	Adopting recommended practices (1 = yes, 0 = otherwise)	+

Source: Author’s computations.

. The logistic probability model [52] is expressed implicitly as:

$${\mathrm{P}}_{\mathrm{i}}(\mathrm{Y}=1/{\mathrm{X}}_{\mathrm{i}})=\mathrm{f}\left({\mathrm{Z}}_{\mathrm{i}}\right)=\frac{1}{{1+\mathrm{℮}}^{-\left(\mathsf{\alpha }+{\mathsf{\beta }}_{\mathrm{i}}{\mathrm{X}}_{\mathrm{i}}+{\mathsf{\epsilon }}_{\mathrm{i}}\right)}}$$

Explicitly, the logistic probability model is expressed as:

$$\begin{array}{cc}\hfill \mathrm{L}\mathrm{o}\mathrm{g}\mathrm{i}\mathrm{t}\left(\frac{{\mathrm{P}}_{\mathrm{i}}}{1-{\mathrm{P}}_{\mathrm{i}}}\right)={\mathrm{Z}}_{\mathrm{i}}& \\ & ={\mathsf{\beta }}_0+{\mathsf{\beta }}_1\mathrm{L}\mathrm{C}\mathrm{T}\mathrm{N}+{\mathsf{\beta }}_2\mathrm{W}\mathrm{U}\mathrm{A}+{\mathsf{\beta }}_3\mathrm{I}\mathrm{T}+\dots {\mathsf{\beta }}_4\mathrm{C}\mathrm{R}\mathrm{C}+{\mathsf{\beta }}_2\mathrm{E}\mathrm{D}\mathrm{U}+{\mathsf{\beta }}_2\mathrm{G}\mathrm{E}\mathrm{N}\mathrm{D}\hfill \\ & +{\mathsf{\beta }}_2\mathrm{T}\mathrm{W}\mathrm{U}+{\mathsf{\beta }}_2\mathrm{A}\mathrm{I}\mathrm{W}+{\mathsf{\beta }}_2\mathrm{I}\mathrm{W}\mathrm{U}+{\mathsf{\beta }}_2\mathrm{F}\mathrm{E}\mathrm{X}\mathrm{P}+{\mathsf{\beta }}_2\mathrm{R}\mathrm{E}\mathrm{C}\mathrm{P}\hfill \end{array}$$

(7)

2.3.4. Method Used to Measure Irrigation Water Accessibility

Four features of access were measured when building the scheme model: equity, adequacy, reliability, and satisfaction. Adequacy and reliability capture the system attributes that permit water distribution to the farmers. Equity represents institutional mechanisms for water distribution, and satisfaction captures the social aspect of the farmers. To test the reliability of the indicators (equity, adequacy, reliability, and satisfaction), a reliability test was conducted, obtaining an acceptable Cronbach’s Alpha of 0.75 and proving internal consistency. The weighted average was obtained for water access across all the locations (upstream, middle, and downstream). The weightings were derived by asking the farmers to score their responses on a 4-point Likert scale from “Strongly Agree” to “Strongly Disagree”. The average score of 9.15 was used to classify the access to water into high and low categories. Farmers below the average score were considered to have low access, while those above the average score were classified as having high access.

3. Results and Discussion

3.1. Irrigation Water Used, Yield per Acre, and Water Productivity

The results in

Table 3. Distribution of respondents based on irrigation water used, yield per acre, and water productivity.

Location	Minimum	Maximum	Mean	Std. Deviation
Irrigation Water Used (m3)
Upstream	3041.98	9777.78	5258.73	2001.69
Middle	1498.51	5649.38	3654.14	1216.26
Downstream	1100.00	5703.70	2429.83	881.27
Yield of Paddy/Acre (kg)
Upstream	1800	3000	2296.51	400.43
Middle	1125	2400	1978.51	343.81
Downstream	1050	2475	1625.25	265.35
Physical Water Productivity
Upstream			0.44 kg/m3
Middle			0.53 kg/m3
Downstream			0.66 kg/m3

Source: Authors’ Computations. Physical water productivity (PWP) = ${fx}^{-1}$, where f is the output produced in kg and $x$ is the quantity of water applied in M³.

indicate that upstream farms have an average irrigation water use of 5258 m³ per acre, contrasting with farms located in the middle and downstream, with 3654.14 m³ and 2429.83 m³, respectively. This result further indicates that upstream farms obtained more yield per acre than the farms located in the middle and downstream. The mean output of the downstream farms was approximately 71% of the mean output of the upstream farms. Likewise, the mean yield of the farms in the middle was 86% of the output of the upstream farms, indicating that the middle and downstream farmers were comparatively worse off. Table 3 also shows that the downstream farmers utilized water more productively (0.66 kg/m³) than their counterparts. Conversely, the upstream farms had the least water productivity (0.44 kg/m³), signaling an overutilization of water. This result has two implications: while the upstream farmers have a significantly higher output than the downstream farmers, the downstream farmers are more efficient in their water use. This finding is consistent with reference [3], who reported that downstream users are constantly neglected in irrigation water availability. In comparison, upstream users have a natural advantage in water extraction, and they continue to overuse and waste water at the expense of downstream farmers.

3.2. Access to Irrigation Water

The results presented in Figure 3 show the access to irrigation water among rice farmers in the upstream, middle, and downstream areas. The result shows that the farmers located downstream had the lowest access to irrigation water, as indicated by all the water accessibility indicators (reliability, equity, adequacy, and satisfaction) used in the analysis.

3.3. Kruskal–Wallis Test

The Kruskal–Wallis test was used to test the hypothesis that all the variables (irrigation water use, access to irrigation water, and quantity of paddy produced) were the same among the categories of farmers located upstream, middle, and downstream. The results in

Table 4. Differences in water use and output among user categories.

Test Variables	Chi-Square	DF	Sig
Irrigation water use (upstream, middle, and downstream)	95.461	2	0.000 ***
Access to irrigation water (upstream, middle, and downstream)	111.164	2	0.000 ***
Paddy output (upstream, middle, and downstream)	72.006	2	0.000 ***

Source: Authors’ computations. *** significant at 1 percent.

show that the tested variables were statically different from one another. However, the analysis does not explain the relationships between the variables. Hence, post-hoc tests were used to test the significance of the variables in each case.

3.4. Post-Hoc Test

A post-hoc test was used to establish whether the groups were distinct from one another.

Table 5. Post-hoc test.

Sample Comparison	T-Statistic	Std. Error	Std. Test Statistic	Sig	Adj. Sig
Irrigation Water Use
Downstream and middle	−43.995	10.161	−4.330	0.000	0.000 ***
Downstream and upstream	−99.475	10.199	−9.754	0.000	0.000 ***
Middle and upstream	−55.480	10.235	−4.421	0.000	0.000 ***
Water Accessibility
Downstream and middle	45.964	10.032	4.582	0.000	0.000 ***
Downstream and upstream	108.101	10.286	10.510	0.000	0.000 ***
Middle and upstream	62.137	10.111	6.146	0.000	0.000 ***
Paddy output
Downstream and middle	−4.167	10.153	−0.410	0.681	1.00 NS
Downstream and upstream	0.77.164	10.329	−7.471	0.000	0.000 ***
Middle and upstream	0.72.996	10.074	−7.246	0.000	0.000 ***

Source: Authors’ calculations. NS = Nonsignificant, *** significant at 1 percent.

shows a significant difference in irrigation water use between the upstream and middle farmers at a 1% probability level. The results also show that the upstream farmers used more water than the downstream farmers at a 1% probability level. These findings showcase that the downstream farmers used less irrigation water than their upstream counterparts. The upstream farmers had a spatial benefit for appropriating more water. Additionally, the results show a statistical difference in the irrigation water use between the middle and upstream farmers.

Regarding water accessibility, the post-hoc test showed a significant difference between the farmers in the entire locations at a probability level of 1%. This result validates the descriptive statistics (Figure 1). The result is also consistent with [26,53], who reported that downstream farmers face more uncertainty and risk because they receive less water than upstream farmers. Reference [54] also claimed that upstream farmers consume more water than downstream farmers, endowing them with comparatively more income, cropping intensity, and yield.

Furthermore, the results (Table 5) showed no significant difference between the paddy output of middle and downstream farmers. However, there was a considerable difference between the downstream and upstream farmers. This is consistent with the findings of [22], who reported a significant difference in wheat yield between upstream and downstream farmers in Punjab, Pakistan. A considerable difference exists between the farmers’ output in the middle and upstream locations of the canal. The crops are mainly stressed during periods of water scarcity, leading to poorer production and poor performance downstream. Low productivity, low cropping intensity, and low profitability indicate significant volatility and revenue instability in their farming.

3.5. Determinants of Paddy Output

A multivariate logistic regression model was generated to predict the probability of the quantity (kg) of rice output per acre produced by the rice farmers in the KRIP. The mean rice output per acre was calculated, and any farmer that had a yield equal to or > the mean was recorded as 1, or otherwise as 0. The independent variables used were the socio-economic and institutional factors (

Table 6. Determinants of paddy output.

Predictor	β	Se β	Wald’s χ2	DF	p. Value	EXP (β)	Tolerance	VIF
Constant	−4.032	2.067	3.804	1	0.051 **	0.018
Location (upstream)	0.937	0.607	4.609	1	0.000 ***	2.939	0.433	2.311
Water user’s association	0.089	0.564	0.025	1	0.875	1.093	0.875	1.143
Illegal tapping	−0.587	0.570	1.061	1	0.303	0.556	0.928	1.077
Condition of the distr. canal	−0.924	0.535	2.980	1	0.084 *	0.397	0.948	1.055
Education	−0.018	0.056	0.099	1	0.754	0.983	0.898	1.113
Gender	−1.883	0.933	4.074	1	0.044 **	0.152	0.903	1.108
Training on water use	0.407	0.694	0.344	1	0.557	1.502	0.863	1.159
Access to irrigation water	0.185	0.075	5.978	1	0.014 **	1.203	0.302	3.309
Irrigation water use	0.977	0.517	3.573	1	0.059 **	2.657	0.348	2.872
Experience	0.001	0.023	0.001	1	0.978	1.001	0.935	1.069
Recommended practices	0.139	0.176	0.625	1	0.429	1.149	0.704	1.142

Notes: *** p < 0.01, ** p < 0.05, * p < 0.10, Chi-square, Cox and Snell R² 0.503, Nagelkerke R² 0.602, −2 Log likelihood 131.211.

). The tested model performed well, with a Cox and Snell R² = 0.502 and Nagelkerke R² = 0.601 and an acceptable significance level of 95 percent. The data revealed that the farmer’s farm location (upstream or downstream) had a statistically significant positive association with their paddy output (exp.β) 2.93. The upstream farmers had a predicted probability of obtaining an output three-times that of their counterparts. The probability of irrigation water use were also statistically significant. The odds ratio of 2.872 showed a positive association between irrigation water use and production, suggesting a direct association between the amount of water applied and paddy output. The probability of access to irrigation water was statistically significant, with a value of (exp. β) 3.309, indicating that better access to irrigation water could increase yields 3.3-fold.

Similarly, the dummy variable of adopting recommended practices was positive and statistically significant, with an odds ratio of 1.15, indicating the probability of its increasing effect on paddy output. This is in line with [55], who reported that adoption of improved and recommended rice production technologies leads to significant yield increase in rice production. Moreover, the dummy variable of training on efficient water use also significantly influenced the farmers’ output (exp.β. 1.5), predicting that training could increase production 1.5 times. This finding also shows that the farmers experienced enhanced chances of increasing production by 0.01 times.

The canals’ physical characteristics were negative and statistically significant (prob. 10%), indicating a detrimental and substantial impact on household paddy output. The famers’ attitudes towards illegal irrigation water tapping and their educational levels were negative, with odds ratios of 1.07 and 1.117, respectively. This indicated that individuals with a greater degree of education had a better earning potential outside of farming and found farming less appealing. This is contrary to the findings of [55], who claimed that education increases the ability of the farmers to adopt agricultural innovation and improve their productivity and efficiency. This finding is comparable to that of [3], who reported that household education is inversely related to access to irrigation water in Sri Lanka.

4. Conclusions

Even though the inequitable distribution of irrigation water is a serious challenge in the KRIP mission of poverty reduction and increasing food production, irrigation is widely pushed in rural Nigeria as an integral approach for increasing food security, income, and livelihoods. This study used a combination of quantitative and physical water measurements to estimate irrigation water usage and test for water accessibility amongst categories of farmers depending on their farm location (upstream, middle, and downstream). Kruskal–Wallis and post-hoc tests were used to evaluate the assumptions that the canal water is evenly distributed and that crop yields (in KRIP) across the location are equal. The determinants of the output were assessed using logistic regression. These findings confirmed the belief that a farmer’s location places them at an advantage or disadvantage due to unequal distribution of irrigation water, which could have a negative impact on downstream farmers’ incomes, food security, and livelihoods.

The quantity of irrigation water use, access to irrigation water, and paddy yield across the locations rejected all the hypotheses. The result of the analysis confirmed that the worst-off farmers are located downstream. Furthermore, the results indicated that the paddy output between the downstream and middle farms was not statistically significant. Farm location, quantity of irrigation water, access to irrigation water, and water usage instructions were all found to increase paddy output.

5. Policy Recommendations

More water for the poor, efficient use of water, and more equitable management are all part of the agricultural water agenda for development. The most effective strategy to alleviate poverty in Sub-Saharan Africa and some parts of Asia and North Africa is to increase overall water productivity though equitable distribution of water resources among the users of large-scale irrigation projects, irrespective of their location (upstream or downstream). Water is the only resource whose scarcity can lead to 100% crop failure. The most pressing policy and intelligent questions for the next few decades will be regarding access to water and water management, which might help to alleviate poverty in the aforementioned regions. Based on the findings of this study, the following policy options are recommended:

The project management should allow farmers, especially those located downstream, to access conjunctive water use. This is very important, as the current project management does not allow farmers to use tube wells in the catchment area. Lining the field canals to reduce seepage and conserve water should be implemented, thereby improving water access for downstream farmers. Fines should be imposed on farmers who break the field canals to acquire more water. Lastly, the management should work closely with the water users’ association for better management of the existing water systems.

The limitation of the study lies in the fact that the study was restricted to only one irrigation project in Nigeria; as such, the result cannot represent the entire situation of irrigation projects in Nigeria. However, it can serve as an eye-opener to investigate similar projects in the country and develop policy options that will improve water accessibility across different locations.