The Effect of Agronomic Management on Micronutrients of Vegetables Grown by Smallholders in Free State and KwaZulu-Natal Provinces of South Africa

Abstract

The production of nutritious food amongst rural farmers has been a challenge for m Cany years. Challenges can be attributed to many factors, including poor access to water, use of old planting methods, financial challenges, etc. Therefore, new climate-smart technologies (CSTs) were introduced to the farmers. The CSTs implemented in the study were in-field rainwater harvesting (IRWH) techniques compared with conventional production (CON). These technologies were applied in combination with sound agronomic management practices, such as mulching and fertilizer application, to produce cabbage, beetroot, spinach and orange-fleshed sweet potato. The vegetables produced were harvested and their nutritional composition analysed to assess whether or not agronomic treatments, water-use technology and season affected their micronutrient levels, with a particular focus on provitamin A and mineral levels. The main finding of the study was that the nutrient levels of the vegetables can be enhanced by adopting IRWH technology combined with different agronomic treatments, especially including mulching as one of the treatment combinations. Limitations: Farmer research participants did not always adhere to research agreements, particularly regarding reserving vegetable samples for analyses. As a result, some experiment replicates are missing. Value: The study findings are of socio-economic significance as they demonstrate that rural, small-scale farmers can apply local, accessible and appropriate agronomic treatments and water-use technologies to achieve economically viable yields of nutritive vegetables to enhance food and nutrition security and household livelihoods of the farmers.

1. Introduction

The discourse on smallholder agriculture in South Africa usually focuses on farmers who produce field crops within the rural villages and how their aspirations, particularly to access agricultural markets, can be realised. Little, if any, attention is given to households who practise gardening, yet they made up to 83.8% of agricultural households in South Africa, as per findings of the 2016 Community Survey [1]. Rural households farm under pronounced resource limitations where dietary diversity in crop choice is likely to be limited and integrated farming is not practised [2]. Low dietary diversity increases the risk of several forms of malnutrition, including micronutrient deficiencies (hidden hunger).

Most household gardens are managed by female household members and their produce, consisting of a combination of leafy and root vegetables, tubers, legumes and other crops, supplements food purchases [3]. Their socio-economic status is similar to that of the traditional smallholder farmer [4] and it is likely that they face similar constraints (e.g., limitations of land size, access to inputs, access to extension services, and water access) and aspirations (i.e., accessing agricultural markets and earning income from their work).

Water access is a major constraint to smallholder agricultural production in most developing countries, including South Africa, where the majority of smallholder agriculture is rain-fed [5]. South Africa is a water-scarce country that receives an average rainfall of 450 mm per year. Many smallholder farmers in South Africa are located in the former homelands, which are characterized as arid and semi-arid and have soils with low agricultural potential [4]. Although smallholder farmers may farm close to rivers, dams and other water sources, they do not have water-use licenses and this restricts them from extracting substantial amounts of water for production [6]. Increasingly, South African households have had to cope with drought, mid-season drought, flooding, etc. and these incidents have been attributed to climate change-related events, sometimes reducing their already limited access to water [7]. These challenging natural conditions and poverty of the farmers make it difficult for them to realize profits from agriculture as the yields are usually low and produce quality, including nutritional quality, is poor.

The farmers in Gladstone (Free State province) and Swayimane (KwaZulu-Natal province) have no access to irrigation; they largely rely on frequently unpredictable rainfall for crop production. The increased occurrence of droughts and floods has further reduced crop yields and quality and could result in the following unfavourable consequences for poor farming households: (i) loss of livelihood security; and (ii) reduction in crop diversity, further lowering dietary diversity [8]. This reduction in dietary diversity could lead to worsening malnutrition in poor households, which cannot afford to replace lost diversity through purchasing vegetables from the market. Diet diversity is important in reducing undernutrition caused by stunting rates as a diverse diet offers more micronutrients needed to address stunting [9]. Reduction in malnutrition is an important SDG (2) goal of improving undernutrition. Adopting climate-smart technologies (CSTs) could improve the capacity of gardening households to respond to these challenges and maintain or increase production levels and improve crop quality properties, possibly positioning themselves to enter the value chain. The exploration of resources of smallholder farmers and how they used them in pursuit of agricultural livelihoods informed the identification of the proposed CSTs for Gladstone (Free State) and Swayimane (KwaZulu-Natal). These proposed CSTs could improve water access among the communities of the areas stated, and the approach used to introduce the CSTs is equally important.

The literature contains numerous examples of policy interventions that failed to achieve their intended goals because the top–down approach was used to introduce technologies to beneficiary communities with little consultation [8]. For undernutrition and hunger (SDG 2) to be improved, policies that bring production and nutrition decisions better are indeed urgent. Informed by this and other developments in the literature, this research proposes the use of a participatory approach to select the technology and use demonstration plots to show its effectiveness. In this research, participatory approaches were adopted because they recognise that communities are knowledgeable of the challenges they face and have possible solutions for the problems. Selected CSTs were demonstrated in various plots in partnership with the communities. By adopting these approaches, the agency and aspirations of the respective communities are also acknowledged, and so are their opinions of the proposed interventions and the reasons why they may or may not succeed [9]. The first step in such a process was gaining stakeholder buy-in, which is important if any intervention is to be successfully adopted. The second step involved showing the farmers evidence that the interventions would address their water access problem. Demonstration plots have been used by agricultural extension officers for decades [10,11]. They have the added advantage of making it possible for the farmers to see how the intervention works and the extension workers can transfer technology to their clients [2].

This study is premised on a need for improved diets for productive farming households and improving prospects of better household wellbeing and health and resultant socio-economic benefits, including improved education outcomes. Good CSTs facilitate the production of diverse crops for diverse diets, and can also introduce farmers to high-value horticultural crops whose surplus can be sold.

The objectives of this study were to determine the effect of water-use technologies and agronomic treatments on nutrient levels, with particular focus on micronutrient levels provitamin A and mineral levels of different vegetable types, and to assess implications for food and nutrition security in the KwaZulu-Natal (KZN) and Free State (FS) provinces.

2. Materials and Methods

Applicable CSTs that complemented the natural and socio-economic conditions in Gladstone (Free State province: 29.3652° S, 26.8395° E) and Swayimane (KwaZulu-Natal province: 29.52302362° S 30.61262580° E; 29.51480680° S 30.65391650° E; 29.53055330° S 30.60564540° E) were implemented and demonstrated in selected homesteads, communities and school gardens to improve production and water use for food and nutrition security at the start-up stage of food value chains. The experiments further determined the effect of agronomic treatments on nutrient levels, specifically provitamin A and mineral levels, of different vegetable types. There is a high prevalence of provitamin A and micronutrients in the study areas, leading to health risks and immunity. Cabbage, beetroot and orange-fleshed sweet potato were planted at Swayimane and at Gladstone. These vegetables were selected because they are major sources of provitamin A and minerals. Further, the communities involved in the study were familiar with planting them and had adopted them as “traditional vegetables”.

2.1. Weather Conditions on Study Sites

Across all demonstration/experiment sites, the nearby representative automatic weather station was used to obtain weather data such as rainfall, which was a major weather parameter required in this study. Vegetables were planted during the 2017/2018 and 2018/2019 growing seasons (between August and February). Early planting and late planting were performed during the 2018/2019 season. In Swayimane, in 2017 and 2018, rainfall during the planting season was 800 mm and 900 mm, respectively, whilst in Gladstone, it was under 600 mm.

2.2. Study Design

Purposive sampling was used to select fields in which experiments were carried out to investigate the effect of water-use technologies and agronomic treatments on the provitamin A and mineral content of different vegetable types. Experimentation applied the complete randomized block design (CRBD) with three replicates for each treatment. The total area for each plot was 36 m².

Conventional tillage (CON) with an application of inorganic fertilizer (Fert) served as the control (CON + Fert), which was compared with the in-field rainwater harvesting (IRWH) technique in combination with various management practices (application of inorganic fertilizer, organic kraal manure incorporated into the soil, and mulching). Inorganic fertilizer, 3:2:1 (28), was applied during planting at an application rate of about 3.57 g (1 bottle top) per planting station. For the treatments with organic fertilizer, manure was collected from the kraal where cattle stayed overnight. The finely dried cattle manure was evenly spread from a 5 L container and incorporated into the soil of every 3 m planting row before planting. This was done to enable the process of decomposition and the release of nutrients to the crop. The quality of the cattle manure used was generally low with less than 3% nitrogen (N), 2% phosphorus (P) and 1% potassium (K) and was influenced by the quality of the rangelands as well as the time of year. For the treatments with mulch, maize stover collected from nearby croplands was applied at an approximate rate of 0.5 kg m⁻² in the basins of the IRWH system. At planting, all the plots were irrigated using a watering can to ensure good seedling establishment. All treatments were cultivated under dryland rain-fed conditions. The treatments included:

• CON + inorganic fertilizer (CON + Fert);
• IRWH + inorganic fertilizer (IRWH + Fert);
• IRWH + mulch (IRWHm);
• IRWH + mulch + inorganic fertilizer (IRWHm + Fert);
• IRWH+ mulch + inorganic fertilizer + manure (IRWHm + Fert + Manure).

Spinach (Spinacea oleracea cultivar Fordhook Giant), beetroot (Beta vulgaris cultivar Detroit Dark Red), cabbage (Brassica oleracea cultivar Drum Head) seedlings and orange-fleshed sweet potato (Ipomoea batatas) cuttings were planted as the test crops.

Weeds were controlled at pre-planting using Roundup^® (glyphosate) herbicide., Swayimane KwaZulu-Natal & Gladstone, Freestate South Africa. During the growing season, mechanical weed control was conducted and problematic pests and diseases were controlled using chemical control methods.

2.2.1. Sample Harvesting and Freeze-Drying

Vegetables are highly perishable during and after harvesting if subjected to unfavourable conditions, especially unsuitable storage conditions. To eliminate these during harvesting, Ziploc bags, cooler boxes and ice cubes were used to preserve the vegetables from the field to the laboratory. The vegetables were harvested randomly from each plot of CSTs. Within each plot, there were three replicates. One beetroot and one cabbage per replicate were randomly selected and harvested. Similarly, one orange-fleshed sweet potato was randomly selected from each replicate and harvested from all CST plots. A bunch of spinach from each replicate was also harvested. Spinach was harvested as above-ground biomass, while beetroot and cabbage were recorded as the mass of total heads per treatment. Furthermore, for the spinach and beetroot, the number of bunches obtained and their mass were recorded per treatment for the determination of economic yield. Sweet potato tubers were harvested and recorded as fresh weight.

Immediately after harvesting, the vegetables were placed in Ziploc bags labelled according to the treatment plot they were harvested from, and the Ziploc bags of each vegetable type were then placed in a cooler box with ice cubes. The vegetables were immediately transported from the field to the laboratory for nutritional analysis.

2.2.2. Sample Preparation and Freeze-Drying

Upon delivery at the laboratory, the vegetable samples were washed with deionised water to remove dirt on their surfaces and were then left to dry at room temperature (approximately 25 °C). Thereafter, each vegetable sample was cut into slices of approximately 100 g and stored in a freezer at −20 °C. The samples were then freeze-dried using a Virtis freeze-dryer (# 6 KBTES-55, SP Industries, 935 Mearns Rd, Warminster, PA 18974, United States USA) at 0.015 kPa, −75 °C for ±5 days until they were completely dry. Finally, the dried samples were milled.

2.3. Nutritional Analysis

Following standard and referenced methods, the raw samples were analysed in triplicates for their proximate composition, mineral elements and provitamin A content. The proximate composition of the vegetable samples was determined according to the methods of the Association of Analytical Chemists [12]. The total mineral content (ash) was determined by the combustion method following the Association of Analytical Chemists (AOAC) Official Method Number 942.05. Fat was analysed by the Soxhlet procedure as described in the AOAC Official Method Number 920.39. Fibre was measured as neutral detergent fibre (NDF) according to the AOAC Official Method Number 2002.0. Protein was determined using the Dumas combustion method described in the AOAC Official Method Number 990.03. The individual mineral elements were determined according to the AOAC Official Method Number 6.1.2 (2002) and inductively coupled plasma (ICP) spectroscopy.

The provitamin A content of the samples was determined by high-performance liquid chromatography (HPLC) using the procedures described in [13].

2.4. Data Analysis

Analysis of variance (ANOVA) was performed on the data using the) to test for significant treatment effects. Statistical significance was evaluated at p < 0.05. The effect of season on the nutritional content was evaluated only for those vegetable types that could be grown in both seasons. Therefore, the effect of season on the provitamin A content was evaluated only for beetroot, while the effect of season on the mineral content was evaluated for both spinach and beetroot.

The rainwater productivity (RWP) that describes the effectiveness with which rainwater is converted into grain, seed or any other edible product by different treatments was calculated by dividing the economic yield harvested from a specific area by the volume of water (in this case, rainfall, as all crops were produced under dryland conditions) the crop received during its growing period. The RWP was expressed as kg ha⁻¹ mm⁻¹ (data not shown] In order to compare treatments, all yields and other relevant parameters (RWP), Ash %; Fat %, Fibre %, Protein %; P %, Zn (mg kg⁻¹), Fe (mg kg⁻¹), and provitamin A (µg kg⁻¹) were expressed in relative units. For each treatment, the average yield obtained from the various sites at each locality was used for the relative yield calculations. For each parameter, the treatment with the highest value has a relative value of 1. Corresponding values of other treatments were then expressed in terms of the treatment with a relative value of 1. For example, if 5000 kg ha⁻¹ was harvested from the IRWH treatment and 3000 kg ha⁻¹ was harvested from the CON treatment, then the relative yield of IRWH was taken as 1 and that of CON would then be 0.6 (3000/5000 = 0.6).

3. Results

As stated earlier, the aim of the study was to investigate the effect of agronomic treatments and water-use technologies on the nutrient levels of different vegetables, with a particular focus on micronutrients, which are predominantly deficient among the study communities, i.e., provitamin A and iron and zinc. The nutrient levels of each vegetable type cultivated under different treatments and water-use technologies in two seasons are presented in

Table 1. Effect of agronomic treatment and water-use technology (either CON or IRWH) on the proximate composition and level (concentration) of individual mineral elements of different vegetables in Swayimane, KwaZulu-Natal province (w/w, on a dry matter basis, mean ± SD).

	Crop	Treatment	Ash (%)	Fat (%)	Fibre (%)	Protein (%)	Ca (%)	P (%)	Zn (mg kg−1)	Fe (mg kg−1)
SEASON 1	Beetroot	CON + Fert	7.74 ± 0.67 c	0.85 ± 0.33 b	12.8 ± 0.60 d	11.52 ± 0.34 a	0.44 ± 0.50 b	0.20 ± 0.17 d	36.00 ± 0.01 b	454.36 ± 0.52 a
		IRWH + Fert	7.89 ± 0.07 c	0.59 ± 0.17 d	13.73 ± 0.32 c	9.82 ± 0.11 b	0.56 ± 0.68 a	0.37 ± 0.47 b	38.21 ± 0.30 a	156.08 ± 0.12 c
		IRWHm + Fert	6.84 ± 0.40 d	0.51 ± 0.23 d	13.45 ± 0.01 c	6.08 ± 0.53 d	0.53 ± 0.60 a	0.38 ± 0.47 b	27.05 ± 0.07 d	136.08 ± 0.11 d
		IRWHm	9.64 ± 0.35 a	1.07 ± 0.00 a	17.51 ± 0.16 a	8.75 ± 0.59 c	0.15 ± 0.44 c	0.45 ± 0.68 a	37.97 ± 0.03 a	232.79 ± 0.28 b
		IRWHm + Fert + Manure	8.52 ± 0.12 b	0.79 ± 0.22 c	14.12 ± 0.40 b	9.02 ± 0.29 b	0.43 ± 0.47 b	0.36 ± 0.44 c	31.12 ± 0.18 c	106.34 ± 0.48 e
	Cabbage	CON + Fert	8.96 ± 0.13 b	1.33 ± 0.39 a	14.59 ± 0.56 b	16.02 ± 0.34 a	0.27 ± 0.09 c	0.71 ± 0.68 a	32.31 ± 0.44 a	92.30 ± 0.42 a
		IRWH + Fert	8.81 ± 0.62 b	1.08 ± 0.18 b	15.73 ± 0.23 a	10.72 ± 0.71 c	0.38 ± 0.20 b	0.42 ± 0.39 b	31.48 ± 0.69 a	57.48 ± 0.68 c
		IRWHm + Fert	8.63 ± 0.08 b	1.17 ± 0.38 b	15.53 ± 0.47 b	13.10 ± 0.03 b	0.25 ± 0.11 c	0.48 ± 0.43 b	19.11 ± 0.15 b	49.00 ± 0.00 e
		IRWHm	10.01 ± 0.48 a	1.42 ± 0.39 a	16.36 ± 0.50 a	13.84 ± 0.66 b	0.97 ± 1.05 a	0.65 ± 0.68 a	19.32 ± 0.46 b	52.31 ± 0.43 d
		IRWHm + Fert + Manure	8.84 ± 0.64 b	1.14 ± 0.04 b	14.57 ± 0.11 b	13.14 ± 0.55 b	0.12 ± 0.41 d	0.13 ± 0.10 c	17.79 ± 0.28 c	63.74 ± 0.36 b
	Spinach	CON + Fert	12.29 ± 0.53 b	3.18 ± 0.16 a	26.96 ± 0.55 b	21.18 ± 0.53 b	0.76 ± 0.12 c	0.43 ± 0.48 b	90.44 ± 0.63 b	1094.30 ± 0.43 c
		IRWH + Fert	15.64 ± 1.31 a	3.07 ± 0.79 a	25.77 ± 0.91 b	21.42 ± 0.97 b	1.75 ± 0.95 a	0.96 ± 1.28 a	115.54 ± 0.77 a	944.50 ± 0.71 d
		IRWHm + Fert	13.71 ± 0.65 b	2.99 ± 0.45 a	24.49 ± 0.54 b	25.50 ± 0.08 a	0.81 ± 0.46 c	0.50 ± 0.51 b	61.25 ± 0.35 c	1600.28 ± 0.40 b
		IRWHm	16.73 ± 0.30 a	3.24 ± 0.67 a	28.41 ± 0.01 a	21.00 ± 0.14 b	0.91 ± 0.19 b	0.40 ± 0.46 b	111.06 ± 0.12 a	1730.08 ± 0.12 a
		IRWHm + Fert + Manure	12.14 ± 0.30 b	2.34 ± 0.02 b	25.29 ± 0.37 b	16.38 ± 0.14 c	0.54 ± 0.29 d	0.28 ± 0.48 c	71.94 ± 0.08 c	1064.95 ± 0.06 a
SEASON 2	Beetroot	CON + Fert	8.74 ± 0.54 b	0.89 ± 0.11 a	17.33 ± 0.20 b	11.82 ± 0.23 b	0.40 ± 0.41 a	0.33 ± 0.31 a	28.20 ± 0.28 b	267.37 ± 0.52 c
		IRWH + Fert	10.11 ± 0.43 a	0.75 ± 0.67 b	20.05 ± 0.51 a	13.72 ± 0.24 a	0.28 ± 0.20 c	0.17 ± 0.04 c	32.11 ± 0.16 a	600.25 ± 0.35 a
		IRWHm + Fert	7.27 ± 0.16 b	0.75 ± 0.39 b	17.73 ± 0.33 b	11.33 ± 0.28 b	0.34 ± 0.29 b	0.18 ± 0.10 b	23.12 ± 0.17 c	357.21 ± 0.30 b
	Spinach	CON + Fert	23.62 ± 0.01 b	2.22 ± 0.01 b	28.62 ± 0.68 b	22.98 ± 0.22 b	0.87 ± 0.11 b	0.53 ± 0.31 c	27.42 ± 0.59 b	557.18 ± 0.25 c
		IRWHm	26.17 ± 0.35 a	3.06 ± 0.51 a	31.41 ± 0.13 a	24.11 ± 0.52 a	1.07 ± 0.54 a	0.63 ± 0.55 b	26.25 ± 0.35 c	1462.10 ± 0.14 a
		IRWHm + Fert	23.36 ± 0.26 b	2.60 ± 0.54 b	24.84 ± 0.38 c	20.33 ± 0.11 c	1.14 ± 0.55 a	0.74 ± 0.68 a	29.31 ± 0.43 a	659.25 ± 0.35 b
	OFSP	IRWH + Fert	3.62 ± 0.01 a	2.16 ± 0.39 a	7.47 ± 0.69 a	5.82 ± 0.04 b	0.24 ± 0.20 a	0.18 ± 0.24 b	6.32 ± 0.45 a	74.30 ± 0.43 a
		IRWHm + Fert + Manure	4.18 ± 0.37 a	1.09 ± 0.19 b	6.53 ± 0.09 b	7.41 ± 0.54 a	0.16 ± 0.17 a	0.02 ± 0.08 c	3.42 ± 0.59 b	7.16 ± 0.22 d
		IRWHm + Fert	4.00 ± 0.69 a	0.86 ± 0.04 c	6.48 ± 0.04 b	6.29 ± 0.03 a	0.24 ± 0.27 a	0.02 ± 0.07 c	3.06 ± 0.09 b	19.00 ± 0.07 b
		IRWHm + Fert + Manure	4.14 ± 0.65 a	1.29 ± 0.38 b	6.91 ± 0.45 b	6.89 ± 0.66 a	0.31 ± 0.24 b	0.24 ± 0.36 a	3.47 ± 0.67 b	12.48 ± 0.67 c

CON = conventional tillage; IRWH = in-field rainwater harvesting; Fert = inorganic fertilizer; Manure = organic fertilizer (kraal manure); m = mulch. OFSP= orange-fleshed sweet potato. For each vegetable type for each season, means marked with different letters in a column are significantly different (p < 0.05), according to the Turkey test.

. As already stated, the treatment combination of conventional tillage with fertilizer application (CON + Fert) served as the control. There were challenges with collecting data from all the experiments or from replicates of some of the experiments because the study participants (smallholder farmers) did not always honour the study arrangements made with the researchers. In a number of instances, the farmers harvested the vegetables and consumed or sold them before the researchers could collect data. The nutrient data for Swayimanein, KwaZulu-Natal (KZN), had much better replication than the data for Gladstone, Free State (FS) province. Therefore, only nutrient data from Swayimane are reported in Table 1 where data only from Swayimane is used, it is explained. Data from both Swayimane and Gladstone were used to calculate relative values for yields and nutrient concentration and then the trends of relative values as illustrated in Figure 1.

3.1. Effect of Water-Use Technology and Agronomic Treatment on the Total Mineral Content (Ash) and Iron and Zinc Concentrations of Different Vegetables

Beetroot

In the first season, IRWH combined with mulching (IRWHm) resulted in a higher concentration of total mineral content (ash) (9.64%) in beetroot relative to the control (CON + Fert) (ash concentration in beetroot = 7.74%) as well as the treatments combining IRWH with the different agronomic treatments (ash concentration ranged from 6.84 to 8.52%) (p < 0.05) (Table 1). However, the control beetroot had a much higher iron concentration (454.36 mg kg⁻¹) than all the beetroot samples produced under experimental conditions (treatment combinations); iron concentration ranged from 106.34 mg kg⁻¹ to 232.79 mg kg⁻¹. In the second season, IRWH in combination with inorganic fertilizer (IRWH + Fert) resulted in a significantly higher concentration of ash (10.11%), zinc (32.11 mg kg⁻¹) and iron (600.25 mg kg⁻¹) relative to the control, whose zinc and iron concentrations were 28.20 mg kg⁻¹ and 267.37 mg kg⁻¹, respectively (Table 1).

Cabbage

Cabbage was planted in Season 1 only (Table 1). A significantly higher concentration of ash (10.01%) was observed in cabbage cultivated under IRWH combined with mulching (IRWHm) relative to the control (8.96%) and other experimental treatments (ash concentration ranged from 8.63 to 8.84%). On the other hand, generally, the control cabbage had significantly higher zinc and iron concentrations compared to the experimental vegetable samples.

Spinach

The results in Table 1 indicate that the IRWHm treatment significantly increased the ash, zinc and iron concentrations of the spinach. In Season 1, spinach produced under IRWHm had an ash concentration of 16.73% compared to a 12.29% ash concentration in the control. Similarly, in Season 2, the ash concentration of spinach produced under IRWHm mulching was 26.17% compared to a 23.62% ash concentration in the control. Further, in Season 1, spinach produced under IRWHm had an iron concentration of 1600.28 mg kg⁻¹ compared to 944.50 mg kg⁻¹ in the control; zinc concentration was 111.06 mg kg⁻¹ compared to 90.44 mg kg⁻¹ in the control (Table 1).

Orange-fleshed sweet potato

The total mineral contents (ash) of sweet potatoes grown under the different experimental combinations of IRWH and agronomic treatments (mulching and/or fertilizer application) were not significantly different (Table 1). Interestingly, however, IRWH + Fert resulted in significantly higher zinc (6.32 mg kg⁻¹) and iron (74.30 mg kg⁻¹) concentrations in sweet potatoes compared to the effect of other experimental treatments, which resulted in zinc and iron concentrations in sweet potatoes ranging from 3.06 mg kg⁻¹ to 3.47 mg kg⁻¹ and 7.16 mg kg⁻¹ to 19.00 mg kg⁻¹, respectively (Table 1).

3.2. Nutrient Reference Values and Dietary Reference Intakes

Table 2. The reference mean values for total mineral content (ash) and iron, zinc and provitamin A concentrations in different types of vegetables [14].

Vegetable Type	Ash (g 100 g−1) (%)	Iron (mg kg−1)	Zinc (mg kg−1)	Provitamin A (µ g−1) (β-Carotene Equivalents)
Cabbage	0.6 (7.5)	5.0 (63.0)	1.5 (18.8)	0.07 (0.88)
Beetroot	1.1 (9.2)	8.0 (67.0)	2.9 (24.0)	0.03 (0.25)
Spinach	2.0 (3.1)	44.0 (550.0)	7.3 (90.1)	4.68 (58.50)
CFSP	0.8 (4.4)	3.0 (16.0)	1.8 (9.9)	0.03 (0.16)
ORFSP	2.3 (12.6)	24.0 (132.0)	9.8 (54.0)	4.56 (25.07)

CFSP, cream-fleshed sweet potato; ORFSP, orange-fleshed sweet potato.

presents reference mean values for total mineral content (ash) and iron, zinc and provitamin A concentrations of different types of vegetables, and

Table 3. Dietary Reference Intakes (IOM, 2016).

Life Stage	Iron (mg day−1)	Zinc (mg day−1)	Vitamin A (μg day−1)
	EAR	EAR	EAR
1–3 years	3.0	2.5	210
4–8 years	4.1	4.0	275
Males: 9–13 years	5.9	7.0	445
Males: 14–18 years	7.7	8.5	630
Males: 19–30 years	6.0	9.4	625
Males: 31–50 years	6.0	9.4	625
Males: 51–70 years	6.0	9.4	625
Males: >70 years	6.0	9.4	625
Females: 9–13 years	5.7	7.0	420
Females: 14–18 years	7.9	7.3	485
Females: 19–30 years	8.1	6.8	500
Females: 31–50 years	8.1	6.8	500
Females: 51–70 years	5	6.8	500
Females: >70 years	5	6.8	500

EAR: Estimated average requirement.

shows dietary reference intakes for comparison with the values obtained in the current study.

It seems that there are few or no studies that have been conducted using the same methodology as the present study. The closest study found was on traditional leafy vegetables (TLVs) and focused on changes in the levels of zinc and provitamin A (beta-carotene) [15]. However, because of differences in the designs of the study reported in [15] and the current study, comparisons cannot be made.

Generally, the study results indicate that, for all the types of vegetables investigated, IRWH combined with different agronomic treatments enhanced nutrient content, including total mineral content (ash) and individual minerals contents, such as zinc and iron, of the vegetables. The ash content of each vegetable type, namely beetroot, cabbage, spinach and sweet potatoes (produced under experimental conditions) ranged from 6.84–10.11%, 8.63–10.01%, 12.14–26.17% and 3.62–4.18% (on a dry matter basis), respectively (Table 1). The maximum value of the range was generally higher than the reference value: beetroot, 9.2%; cabbage, 7.5%; spinach, 3.1%; and sweet potato, 4.4% (on a dry matter basis) (Table 2). The zinc and iron values showed the same trend as the ash values (Table 1 and Table 2).

Vegetables grown under certain conditions may contain too high concentrations of minerals, especially trace minerals such as zinc and iron, which may pause a health risk to consumers. Therefore, the iron and zinc values of the vegetables presented in Table 1 were compared with dietary reference intakes presented in Table 3. From Table 1 and Table 3, it can be deduced that the mineral concentrations of the vegetable samples would contribute significantly to achieving the estimated average requirement (EAR) of iron and zinc for the different population groups included in Table 3. Hypothetically, if an individual from any of the population groups consumed an exceptionally large amount of any of the vegetable types included in the present study such that 100 g of dry matter of that vegetable type is ingested, per day, he/she would achieve more or less the EAR for iron and zinc, respectively, without any risk of mineral toxicity.

3.3. Effect of Season on the Total Mineral Content (Ash), Iron and Zinc Levels of Different Vegetables

Results in

Table 4. Effect of season, agronomic treatments and water-use technology on the proximate composition of different vegetables in Swayimane, KwaZulu-Natal province (g 100 g⁻¹, dry matter basis).

Crop	Treatment		Mean ± SD
			Ash		FAT		NDF		Protein
			First Season	Second Season	First Season	Second Season	First Season	Second Season	First Season	Second Season
Beetroot	CON + Fert		7.74 ± 0.67	7.56 ± 0.18	0.84 ± 0.33	0.61 ± 0.52	12.82 ± 0.6	15.83 ± 0.48	11.53 ± 0.35	13.44 ± 0.69
	IRWH + Fert		7.81 ± 0.07	10.11 ± 0.43	0.59 ± 0.18	0.76 ± 0.67	13.73± 0.32	20.06 ± 0.52	9.82 ± 0.12	13.72 ± 0.24
	IRWHm + Fert		6.84 ± 0.4	7.27 ± 0.16	0.51 ± 0.23	0.76 ± 0.39	13.45± 0.01	17.73 ± 0.34	6.09 ± 0.53	11.33 ± 0.28
	Mean Square		0.59	4.88	0.06	0.02	0.43	8.98	15.48	3.41
	p value	T	0.000		0.000		0.000		0.000
		S	0.000		0.000		0.000		0.000
		TxS	0.000		0.000		0.000		0.000
Spinach	CON + Fert		12.29 ± 0.53	23.62 ± 0.01	3.19 ± 0.16	2.23 ± 0.01	26.96 ± 0.55	28.63 ± 0.69	21.19 ± 0.54	22.98 ± 0.23
	IRWH + Fert		15.64 ± 1.31	26.18 ± 0.35	4.07 ± 0.79	3.06 ± 0.51	25.78 ± 0.91	31.42 ± 0.13	21.43 ± 0.97	24.11 ± 0.52
	IRWHm + Fert		13.71 ± 0.65	23.37 ± 0.26	3 ± 0.45	2.61 ± 0.54	24.49 ± 0.54	24.84 ± 0.38	25.51 ± 0.08	20.33 ± 0.11
	Mean Square		5.66	4.84	0.65	0.35	3.05	21.81	11.79	7.53
	p value	T	0.045		0.010		0.001		0.021
		S	0.000		0.045		0.000		0.000
		TxS	0.087		0.411		0.132		0.001

and

Table 5. Effect of season, agronomic treatments and water-use technology on the mineral levels of different vegetables in Swayimane, KwaZulu-Natal province (w/w, on dry matter basis).

Crop	Treatment		Mean ± SD
			Ca (%)		P (%)		Zn (mg kg−1)		Fe (mg kg−1)
			First Season	Second Season	First Season	Second Season	First Season	Second Season	First Season	Second Season
Beetroot	CON + Fert		0.45 ± 0.50	0.41 ± 0.42	0.20 ± 0.18	0.33 ± 0.31	36.01 ± 0.01	28.20 ± 0.28	454.37 ± 0.52	267.37 ± 0.52
	IRWH + Fert		0.56 ± 0.68	0.29 ± 0.21	0.38 ± 0.48	0.18 ± 0.05	38.22 ± 0.31	32.12 ± 0.16	156.09 ± 0.12	600.25 ± 0.35
	IRWHm + Fert		0.54 ± 0.61	0.35 ± 0.29	0.38 ± 0.47	0.18 ± 0.10	27.05 ± 0.08	23.12 ± 0.17	136.08 ± 0.12	357.22 ± 0.30
	Mean Square		0.01	0.01	0.02	0.02	69.98	40.72	63559.96	59315.23
	p value	T	0.000		0.000		0.000		0.000
		S	0.000		0.000		0.000		0.000
		TxS	0.000		0.000		0.000		0.000
Spinach	CON + Fert		0.77 ± 0.13	0.66 ± 0.06	0.43 ± 0.49	0.32 ± 0.18	90.45 ± 0.63	27.42 ± 0.59	1094.30± 0.43	557.18 ± 0.25
	IRWH + Fert		1.75 ± 0.95	1.08 ± 0.54	0.97 ± 1.28	0.63 ± 0.55	115.55 ± 0.77	26.25 ± 0.35	944.51 ± 0.71	1462.10 ± 0.14
	IRWHm + Fert		0.82 ± 0.46	1.14 ± 0.55	0.51 ± 0.52	0.75 ± 0.69	61.25 ± 0.36	29.31 ± 0.44	1600.29± 0.41	659.25 ± 0.35
	Mean Square		0.61	0.14	0.17	0.10	1477.05	4.77	236170.11	491288.87
	p value	T	0.000		0.124		0.000		0.000
		S	0.000		0.000		0.000		0.000
		TxS	0.000		0.258		0.000		0.000

CON = conventional tillage; IRWH = in-field rainwater harvesting; Fert = inorganic fertilizer; m = mulch; T = treatment; S = season.

indicate that agronomic treatment and water-use technology are not the only factors that affected the nutrient content of the vegetables; season had an effect as well. The effect of season on the nutrient content (other than provitamin A, which is dealt with later) of vegetables was investigated in beetroot and spinach. Generally, beetroot and spinach produced in the second (dry) season had higher nutrient content than the samples of the same respective vegetable type produced in the first season (wet season). The results suggest that vegetables produced in the dry season have higher nutrient concentration relative to their nutrient concentration when produced in the first season (wet season). However, these results may be attributed to dilution of nutrients in the vegetable samples of the wet season because the nutrients were distributed across a larger biomass (yield) in the first (wet) season relative to the biomass obtained in the second (dry) season. As explained previously, Swayimane had much better replication than the data for Gladstone, Free State (FS) province. Therefore, only data from Swayimane are reported in Table 4.

3.4. The Effect of Agronomic Treatment, Water-Use Technology and Season on the Provitamin A Content of Different Vegetable Types

Table 6. Effect of agronomic treatment and water-use technology on the carotenoid content of different vegetables in Swayimane, KwaZulu-Natal province (µg g⁻¹, dry basis).

	Crop	Treatment	Mean ± SD
			Lutein	Zeaxanthin	β-Cryptoxanthin	13-cis-BC	BC	9-cis-BC	Provitamin A Content
Season 1	Cabbage	CON + Fert	0.040 ± 0.002 e	0.178 ± 0.005 e	0	0.233 ± 0.011 a	0.239 ± 0.011 a	0.233 ± 0.011 a	0.704 ± 0.034 a
		IRWH + Fert	0.090 ± 0.001 a	0.381 ± 0.004 b	0	0.223 ± 0.001 a	0.234 ± 0.000 a	0.220 ± 0.001 a	0.676 ± 0.001 a
		IRWHm + Fert	0.084 ± 0.002 b	0.394 ± 0.028 a	0	0.240 ± 0.023 a	0.248 ± 0.024 a	0.240 ± 0.023 a	0.727 ± 0.070 a
		IRWHm	0.072 ± 0.003 c	0.309 ± 0.001 c	0	0.218 ± 0.001 a	0.232 ± 0.002 a	0.218 ± 0.001 a	0.668 ± 0.005 a
		IRWHm + Fert + Manure	0.061 ± 0.002 d	0.268 ± 0.005 d	0	0.215 ± 0.005 a	0.225 ± 0.006 a	0.215 ± 0.005 a	0.655 ± 0.015 a
	Beetroot	CON + Fert	0.064 ± 0.001 a	0.289 ± 0.010 a	0	0.205 ± 0.001 a	0.237 ± 0.000 a	0.204 ± 0.000 a	0.646 ± 0.001 a
		IRWH + Fert	0.035 ± 0.001 e	0.192 ± 0.005 c	0	0.201 ± 0.004 a	0.211 ± 0.003 a	0.199 ± 0.004 a	610 ± 0.011 a
		IRWHm + Fert	0.042 ± 0.001 d	0.214 ± 0.025 b	0	0.221 ± 0.014 a	0.248 ± 0.016 a	0.222 ± 0.015 a	0.691 ± 0.045 a
		IRWHm	0.050 ± 0.005 c	0.214 ± 0.021 b	0	0.237 ± 0.026 a	0.267 ± 0.031 a	0.237 ± 0.028 a	0.741 ± 0.085 a
		IRWHm + Fert + Manure	0.053 ± 0.006 b	0.158 ± 0.002 d	0	0.216 ± 0.005 a	0.226 ± 0.005 a	0.217 ± 0.005 a	0.659 ± 0.014 a
Season 2	Beetroot	CON + Fert	0.140 ± 0.012 b	0.637 ± 0.050 b	0.132 ± 0.001 a	0.304 ± 0.000 a	0.581 ± 0.003 a	0.307 ± 0.008 a	1.258 ± 0.004 a
		IRWH + Fert	0.160 ± 0.008 a	0.868 ± 0.072 a	0.127 ± 0.005 b	0.280 ± 0.004 b	0.556 ± 0.014 b	0.303 ± 0.002 a	1.203 ± 0.014 a
		IRWHm + Fert	0.098 ± 0.004 c	0.443 ± 0.029 c	0.111 ± 0.005 c	0.231 ± 0.004 c	0.325 ± 0.005 c	0.226 ± 0.006 b	0.838 ± 0.018 b
	OFSP	IRWH + Fert	0.122 ± 0.013 a	0.626 ± 0.067 a	0.163 ± 0.005 b	6.321 ± 0.229 a	30.789 ± 0.889 a	0.977 ± 0.080 a	38.169 ± 0.738 a
		IRWHm + Fert	0.041 ± 0.002 b	0.281 ± 0.008 b	0.403 ± 0.052 a	2.127 ± 0.009 b	14.041 ± 0.516 b	0.490 ± 0.014 b	16.860 ± 0.537 b
		IRWHm + Fert	0.000 ± 0.000 c	0.143 ± 0.005 d	0.135 ± 0.010 c	0.664 ± 0.040 d	3.568 ± 0.228 d	0.272 ± 0.006 d	4.571 ± 0.279 d
		IRWHm + Fert + Manure	0.040 ± 0.001 b	0.252 ± 0.008 c	0.136 ± 0.002 c	1.823 ± 0.064 c	12.393 ± 0.013 c	0.450 ± 0.012 c	14.734 ± 0.064 c
	Spinach	CON + Fert	23.290 ± 0.170 d	182.941 ± 17.581 c	0	8.411 ± 0.877 b	53.157 ± 2.061 b	10.547 ± 0.359 b	72.115 ± 3.298 b
		IRWH + Fert	24.242 ± 0.403 c	152.804 ± 3.085 d	0	5.187 ± 0.181 d	29.440 ± 1.900 e	6.840 ± 0.405 e	41.466 ± 2.485 e
		IRWHm + Fert	25.842 ± 0.473 b	191.179 ± 1.618 b	0	13.383 ± 0.090 a	100.787 ± 0.954 a	21.007 ± 0.040 a	135.178 ± 0.903 a
		IRWHm	21.725 ± 0.140 e	144.776 ± 4.689 e	0	6.393 ± 0.046 c	38.879 ± 1.258 d	7.093 ± 0.951 d	52.364 ± 2.163 d
		IRWHm + Fert + Manure	33.958 ± 0.117 a	207.975 ± 0.416 a	0	8.455 ± 0.166 b	49.022 ± 1.249 c	9.666 ± 0.845 c	67.143 ± 2.260 c

BC = β-carotene; Total provitamin A (β-carotene equivalents) = (0.5β-cryptoxanthin µg/g + 13-cis-BC + BC + 9-cis-BC). CON = conventional tillage; IRWH = in-field rainwater harvesting; Fert = inorganic fertilizer; Manure = organic fertilizer (kraal manure); m = mulch. OFSP = orange-fleshed sweet potato. For each vegetable type for each season, means marked with different letters in a column are significantly different (p < 0.05), according to the Turkey test.

shows the effect of agronomic treatment and water-use technology on the provitamin A content of various vegetables. For all the different types of vegetables studied, except beetroot, the effect of agronomic treatment and water-use technology was not tested over the two seasons; therefore, treatment effects were assessed within one season. Water-use technology and agronomic treatment had no effect on provitamin A content. For sweet potato, the total provitamin A content was highest when the vegetable was produced under IRWH in combination with inorganic fertilizer (38.169 µg g⁻¹) compared with the total provitamin A content ranging from 4.571 µg g⁻¹ to 16.860 µg g⁻¹ for sweet potatoes produced under IRWH in combination with other agronomic treatments. As stated earlier, data for the control could not be collected for the sweet potato experiments. The combination of IRWH with organic and inorganic fertilizers resulted in the spinach sample having a much higher total provitamin A content (135.178 µg g⁻¹) compared to the control (72.115 µg g⁻¹) and other spinach samples that were produced under other experimental conditions (their total provitamin A content ranged from 41.466 µg g⁻¹ to 52.364 µg g⁻¹) (Table 6).

Among the different types of vegetables investigated in the study, only beetroot was planted in both seasons (Table 6). Therefore, the effect of season on the provitamin A content of vegetables was determined for beetroot only (

Table 7. The effect of season, agronomic treatment and water-use technology on the carotenoid content of beetroot in Swayimane, KwaZulu-Natal province (µg g⁻¹, dry basis).

Treatment		Mean ± SD
		Lutein		Zeaxanthin		Provitamin A
		First Season	Second Season	First Season	Second Season	First Season	Second Season
CON + Fert		0.064 ± 0.001 b	0.289 ± 0.010 b	0.140 ± 0.012 b	0.637 ± 0.050 b	1.258 ± 0.004 a	0.646 ± 0.001 b
IRWH + Fert		0.160 ± 0.008 a	0.868 ± 0.072 a	0.160 ± 0.008 a	0.868 ± 0.072 a	1.203 ± 0.014 a	0.610 ± 0.011 b
IRWHm + Fert		0.053 ± 0.006 b	0.214 ± 0.025 b	0.098 ± 0.004 b	0.443 ± 0.029 b	0.838 ± 0.018 b	0.691 ± 0.045 a
p value	T	0.000		0.000		0.045
	S	0.000		0.000		0.000
	TxS	0.231		0.0590		0.030

CON = conventional tillage; IRWH = in-field rainwater harvesting; Fert = inorganic fertilizer; m = mulch; T = treatment; S = season. Mean values (in the same column) marked with different letters (a, b) are significantly different (p < 0.005).

). Planting season significantly affected the carotenoid content, including provitamin A, of beetroot. Overall, higher levels of lutein, zeaxanthin and provitamin A were obtained from beetroot cultivated in the second (dry) season. For example, when beetroot was produced under IRWH + Fert, its provitamin A concentration was 1.203 µg g⁻¹ in the second (dry) season compared to 0.610 µg g⁻¹ in the first (wet) season. Similar to the results of other nutrients (Table 4 and Table 5) discussed earlier, these results (Table 7) may be attributed to the dilution of nutrients in the vegetable samples of the wet season because the nutrients were distributed across a larger biomass (yield) in the first (wet) season relative to the biomass obtained in the second (dry) season. Although the experiments were not uniform among the vegetable types within the same season and across the two seasons, spinach produced under all the conditions, including the control condition, had the highest provitamin A values (Table 6). It is noted that lutein and zeaxanthin are types of carotenoid pigments that do not possess provitamin A activity.

With the exception of cabbage, the maximum value of the range of total provitamin A values obtained for each type of vegetable included in the current study (Table 6) is generally higher than the reference value (Table 2). For example, the highest total provitamin A content of 135.178 µg g⁻¹, which was obtained from the spinach sample grown under IRWH combined with mulching and inorganic fertilizer (IRWHm + Fert), is more than twice the reference value for the total provitamin A content of spinach (58.50 µg g⁻¹) (Table 2). It is encouraging to note that, similar to the case of iron and zinc discussed earlier, if an individual from any of the population groups included in Table 3 consumed an exceptionally large amount of any of the vegetable types included in the present study (except cabbage which was very low in provitamin A) such that 100 g of dry matter of that vegetable type is ingested, per day, the amount of provitamin A ingested would contribute significantly towards achieving the EAR for vitamin A (Table 3), without any risk of vitamin A toxicity.

3.5. Relative Values of Yield and Nutrient Content of Different Vegetable Types Grown in Free State and KwaZulu-Natal under Different Agronomic Treatments and Water-Use Technologies

Figure 1 shows the relative values of yield and nutrient content of different vegetable types grown in Free State (FS) and KwaZulu-Natal (KZN) under different agronomic treatments and water-use technologies. As stated earlier, the treatment combination of CON with the application of inorganic fertilizer (CON + Fert) was used as the control. The results in the figure show that relative to the control, for all the types of vegetables investigated in the two provinces (FS and KZN) IRWH combined with the different agronomic treatments resulted in a higher yield and nutrient content, including micronutrients, i.e., provitamin A and minerals.

3.6. Water Use and Positive Prospects for Food Security and Livelihoods of Smallholder Farmers

As evident in Figure 1, IRWH in combination with mulching and application of inorganic fertilizer and manure (IRWHm + Fert + Manure) is shown to be the best-performing technology in improving yields for the four vegetables, particularly for spinach and cabbage. This is significant because both cabbage and spinach are the most preferred on a daily basis as green leafy vegetables for many households in the study sites and similar settings. They are both provitamin A sources. Improved yields are critical for improving food availability and for income generation, which can be used to purchase more food and expand diet diversity. The improved nutrient content is further important for nutrition access and nutrient availability for households. As established in the literature, micronutrients are critical for improving human development and cognitive development, which impacts adult productivity and positive prospects in learning at school, particularly for young children in households and those in childbearing phases.

4. Conclusions

Except for sweet potato, where treatment had no effect, total mineral content (ash) of the other types of vegetables included in the study was enhanced by combining IRWH with mulching. The results also showed that the zinc and iron levels of the vegetables can be manipulated by combining IRWH with different agronomical treatments, such as mulching and application of fertilizers. The findings also indicate that agronomical treatment, water-use technology and season affect provitamin A content of the vegetables studied. In the second season, where the water was scarce, the provitamin A levels for both spinach and beetroot were significantly higher when cultivated under CON. Overall, the study demonstrates that agronomic treatment, water-use technology and season affect the nutrient levels of different vegetables, including the total mineral content (ash) and the levels of iron, zinc and provitamin A. The improved yields of the IRWH indicated positive prospects for market-based production for small farmers provided other push and pull factors for market access are in place. The improved yields further improve food availability for the household and, thus, improve both availability and utilisation pillars of food and nutrition security. There is a need to conduct more investigations with more vegetable types and treatment combinations to establish the best farming practices using the innovative technologies established in the current study. Other recommendations include that market linkages (both formal and informal markets) be strengthened and established before planting in order to reward the improved yields through sales and improved food availability for the household.