A Review of Controlled Environment Agriculture (CEA) Vegetable Production in Africa with Emphasis on Tomatoes, Onions and Cabbage

Abstract

This chapter reviews the available information about performance indicators for controlled environment agriculture (CEA) and conventional production systems in Africa with an emphasis on those arising from tomatoes, onions and cabbage production. We identified a small number of studies that reported, yields per land area, costs, cumulative energy demand (CED), global warming potential (GWP) and water use for either CEA or field-based production systems. The available information does not allow robust comparisons of CEA and field-based production for any of these indicators, which suggests the need for expanded and improved crop-specific data collection from existing operations and the usefulness of alternative approaches such as economic engineering.

1. Introduction

More than 25% of the world’s population suffers from micronutrient deficiencies and related health problems [1, 2]. Increased vegetable consumption has been proposed as a mechanism to reduce the prevalence of non-communicable diseases (NCD) in low and middle-income countries [1]. However, increasing vegetable consumption faces the challenge of increasing availability (production) at affordable costs [2, 3]. Total conventional (field-based) vegetable production increased in Africa from 2001 to 2021, with both tomatoes (Solanum lycospersicum L.) and onions (Allium cepa L.) production increasing by 59%, and cabbage (Brassica oleracea var. capitata L.) by 65% [4]. Field-based vegetable production in Africa is often practiced close to water supply points, in swampy areas or along the littoral band with easy access to water. In such areas, farmers operate within an informal economy and cultivate plots generally less than 1 hectare in size [3]. Production practices are characterized by the use of limited machinery and other inputs and hand-powered technology such as the use of cutlass, hoe and irrigation boxes [3]. African vegetable farmers generally have access to limited information about technical recommendations [4, 5]. This contributes to a wide range of negative impacts on the environment that include reduction in crop yield and subsequently income and revenue, biodiversity loss, deterioration of water catchments, declining plot sizes, land degradation and greenhouse gas (GHG) emissions [5, 6].

The increased demand for affordable and nutritious food in urban areas has resulted in more demand for land and high migration from rural to urban areas, with workers willing to carry out conventional vegetable farming. Supply chains for vegetables produced in the open fields of rural are usually informal with low levels of coordination [1]. Disruptions in international food supply chains due to COVID-19 and economic and political instability in the region have also compounded the inability to attain regional self-sufficiency in vegetable consumption.

Recent years have seen increased discussion about whether alternatives to field-based vegetable production such as controlled environment agriculture (CEA) provide a mechanism to increase supply in urban areas. CEA comprises multiple types of approaches at alternative scales, including the production of plants, fish, insects, or animals using in- (home production or indoor gardens), medium- (e.g., community gardens), or larger-scale commercial operations, e.g., rooftop greenhouses, plant factories (PF) or vertical farms (VF) often using hydroponics, aquaponics or aeroponics, and growth chambers. These technologies control to varying degrees environmental parameters such as humidity, light, temperature and CO₂ to create optimal growing conditions [6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16]. CEA technologies are classified according to the type of facility and growing systems [16]. Soil-based CEA systems use regular soil or compost as the plant growth medium and the predominant type of greenhouses in Africa [5, 7, 12, 13, 14, 15, 16]. In contrast, hydroponic systems are soilless culture in which solutions containing nutrients are applied directly to the roots of the plants. Aquaponic systems combine fish cultivation and hydroponics plant production. In aquaculture, microbial activity converts fish excreta into nutrients. The nutrient-rich wastewater is then pumped through the hydroponic area for use as the plant nutrient. The plants take up the nutrients and clean the water, which is then cycled back into the fish tank.

There has been significant rise in global CEA production for over the past two decades; the primary crops in CEA production focus on vegetables that form part of the local diet across all income groups (e.g., tomatoes, peppers, spinach, cabbage, and other leafy greens). Whether sold on local markets or to ‘niche’ markets such as hotels, the price is the same as for soil-grown produce [17]. The global–defined here as including vegetables that are important in diets across all income groups (e.g., tomatoes, peppers, spinach, cabbage, and other leafy greens)—has increased by ∼80% from $ 0.4 billion in 2013 to $8.5 billion in 2022, and it is projected to reach about $20 billion in 2026. The main production regions in 2022 were North America ($1.4 billion), Europe ($1.4 billion) and Asia-Pacific ($1.3 billion). Africa and the rest of the world contributed about 14% of the global total, $665 million [17, 18]. CEA operations can often use fewer pesticides, land and water per unit of product [1], and reduce susceptibility to pests, diseases and adverse weather conditions. CEA has the potential to produce year-round yields of high-quality produce with yields as much as 20 times higher [7, 8, 17]. Depending on the technologies compared, CEA can use considerably less water and with the use of renewable energy sources could reduce GHG emissions per kg of product [1, 8]. CEA has received particular attention for production in urban areas where demand is large, but land is limited. CEA may play a beneficial role in the production of vegetables, for which Africa’s per capita consumption is the lowest in the world although demand is increasing. Africa’s vegetable imports amounted to US$ 1.9 billion in 2013. Most of the vegetables imported (tomato, lettuce, and onion) can be produced in the region [17].

In addition to the potential impacts on the supply of vegetables, peri-urban CEA development in low-income counties could provide opportunities for young people migrating to cities from rural areas who are in search of more profitable and less physically demanding work than traditional farming or because rural livelihoods are now less viable [16]. Growers in Kenya, Nigeria and India who provide training in hydroponics or aquaponics identified four common categories of people who are interested in starting commercial CEA ventures: young people looking to start their own business; conventional farmers wanting to try a new approach or boost insufficient income; people in other professions seeking additional income; and white-collar workers who are close to retirement.

Despite the advantages of higher yields and lower use of some inputs [8, 9, 10, 11] some CEA production systems (e.g., plant factories or vertical farms) do not appear to be particularly “climate-smart” in the sense that they can have high production costs, and use more energy and emit more GHG per unit product [8, 9, 10, 11]. CEA activity based primarily on soil-based greenhouses is growing in parts of Africa [16]. Egypt and South Africa have seen development of large-scale greenhouse projects in cooperation with Dutch businesses, greenhouses have a ‘considerable’ presence in Kenya, and a few cases in other countries exists (Nigeria, Namibia, and Somaliland). Efforts to promote increased production of fruits and vegetables with CEA on small farms in low-income countries have achieved limited success. The uptake of CEA technology particularly has been limited in low and middle-income countries, particularly in Africa, with high costs for installation and maintenance cited as constraints [1, 11, 12]. However, distance to market, having government support and access to social media, additional information are key positive determinants of awareness that would help inform the potential role for CEA in Africa (e.g., [19, 20, 21, 22]). To date, there are few empirical studies that carry significant ecological impacts, food insecurity, nutrition-related problems, farmer livelihood challenges, and persistent food system-related inequities on CEA production in Africa [16, 17].

The objective of this chapter is to review the available evidence about the costs and selected environmental performance indicators for CEA and field-based irrigated production systems in Africa with an emphasis on tomatoes (20.7 million tonnes) and onions (15.50 million tonnes) for being some of the predominant vegetables grown in 2021 [4]. Literature search showed up no public data to demonstrate the economic and environmental viability of large-scale CEA production of cabbages and other crops of the cabbage family (4.00 million tonnes) [2, 17].

The evidence is derived from previous studies that reported information about the costs per unit, yields per land area, cumulative energy per unit, GHG emissions an global warming potential (GWP) and water use for CEA and conventional vegetable production. This will provide more information on the potential role for CEA production systems in Africa and highlight the priority needs for additional research.

2. Materials and methods

This review considers six performance indicators per unit of product: cost of output, yields per land area, cumulative energy, GHG and GWP and water use for three crops (cabbage, onions, and tomato) in Africa. The information derives from studies with a Life Cycle Inventory (LCI) and Life Cycle Assessment (LCA) perspectives. LCI and LCA are related in the sense of providing accounting of comprehensive and systematic documentation of the impacts, processes and material flows of production but LCI focuses on current operational inputs and impacts (like energy use in a greenhouse) whereas LCA includes the inputs and impacts of ‘embedded inputs’ (for example, the energy to manufacture and dispose of steel used in a greenhouse). Previous reviews (e.g., [23, 24]) have adopted a similar approach to synthesis of the available information.

We developed a database with information about the six performance indicators using two approaches. For the first approach, we identified literature published from January 2000 to January 2023 using Scopus and Web of Science with search terms ‘life cycle inventory’ AND (‘greenhouse’ OR ‘CEA’) AND (‘tomatoes’) and evaluated which studies had information on production systems in Africa. Fifty-four articles were obtained from the initial search for tomato production. An additional three studies were identified from the literature cited by the articles identified through the database searches. The second approach employed a search in Google scholar for ‘Africa’ AND (‘onions’ OR ‘cabbage’ OR ‘tomatoes’) AND ‘irrigated’ and ‘CEA production’. This search resulted in seven studies on irrigated and CEA production, of which three [15, 16, 25] contained multiple data points (for either for different crops or production systems).

Although 64 studies were reviewed, the number of observations for which specific values of the six indicators per unit of the product were reported or could be calculated was considerably smaller. Many studies reported ranges of values for aggregations of multiple crops (e.g., [15, 25]), which we deemed insufficiently specific for this review. We thus excluded the observations derived from these studies. Our review identified four observations for unit costs per (2 for tomatoes from [26] and two for general vegetables [5, 15]), and two values for each of Cumulative Energy Demand (CED; MJ/kg [26]), Global Warming Potential (GWP; kg CO₂eq/kg [26]) and water use (lts/kg [7, 16, 26]).

One possible reason for the limited number of studies with information for specific crops is the predominance and importance of “mixed vegetable” production systems with multiple crops grown throughout the year. In some sense, it is the overall performance of these systems that is important for the farms producing them, which are often of smaller scale. Thus, studies of the overall performance metrics of costs, yields and water use [15, 26] are more common than studies reporting that information for individual crops. In addition, most studies did not report all metrics of costs, yields, CED, GWP and water use; production indicators were more commonly reported than estimates of energy or GWP. In some cases, we made conversions of available data to estimate appropriate metrics. For example, water use was sometimes reported in units of liters per m² per day, which required an estimate of the length of the growing season in addition to the conversion of yields to kg/m².

The functional unit for our analysis (a measurement that is normalized across all systems for comparative purposes) was 1 kg of product (multiple vegetables or tomato) grown each year. We undertook conversion calculations (e.g., total yields to yields/ha or total GHG to GHG/kg or total water per ha to water/kg product) that were specific to each study. Yields were estimated in kg/m² [7] were converted to hectare by multiplying by 10,000 kg ha⁻¹. We converted values in local currencies to USD using exchange rates at the time data were collected in previous studies.

Revenues ($/ha) were calculated based on product yield (kg/ha) and output price, when available. Total costs of production include both variable and field costs, converted to $/kg when needed using information on yields per hectare and costs per ha. The value of water used was converted to water per kg using volume and yield factors in [26]. Cold greenhouse vegetable was 20.0 kg/m³ yield/water use, where 1 m³ was equivalent to 1000 liters.

Quantities of energy in the standard unit of energy were expressed based on the International System of Units (SI), the joule (symbol J), is equal to 3600 kilojoules or 3.6 MJ. It was converted to megajoules (MJ) based on specific energy density for fuel (36 MJ l_1 in [26]) and later MJ/kg and was adapted for this review.

3. Results and discussion

As noted above, the number of observations with sufficient information for inclusion in this review is small (Table 1). Thus, it is difficult to make direct comparisons of revenues and profitability between operators who are cultivating the same crop in different systems and countries. Some studies [15, 26] provide insights about the impacts on the profitability of alternative irrigation systems for vegetable products, but do not allow comparisons of field and CEA production. More evidence is needed to understand the economic feasibility of CEA vegetable production in low-income countries and the extent to which experience from high-income countries is relevant for low-income countries. That is, no analyzes comparable to [11] have been conducted yet for low- and middle-income countries. Differences in costs and profitability for CEA and conventionally-grown produce may narrow as the initial investment is amortized, productivity increases, and new, cost-effective technologies become available [21, 22, 25, 26] but it is difficult to predict when this might occur.

Products, production system	Production cost	Cumulative energy demand (CED; kWh/kg)	Global warming potential (GWP; kg CO2eq/kg)	Water usage	Total
Tomatoes
Greenhouse	1	1	1	1	4
Open field	1	1	1	1	4
Multiple vegetables
Greenhouse	0	0	0	0	0
Open field	2	0	0	1	3
Total
Greenhouse	1	1	1	1	4
Open field	3	1	1	2	7

Table 1.

Summary of observations from review.

There is anecdotal evidence that hydroponic systems may be more economically viable for vegetable production, including in dry land climates due to their minimal water use [https://bicfarmsconcepts.com]. In these systems, inexpensive, locally-available materials were used as substrates [23, 25, 26]. Where electricity was expensive or its supply irregular, pumps that do not need to lift or spray water on to the roots, such as the gravity-driven Kratky or ebb and flow technologies may be preferred. Where there is sufficient water and reliable electricity, aquaponics can be a viable form because it has two outputs—vegetables and fish—that provide complementary sources of income [26].

There was some evidence that enclosed structures using shipping containers and re-purposed buildings can house financially, socially and environmentally viable. CEA operations, partly because they have enabled entrepreneurs to set up in built-up urban locations where there is no space for greenhouses [16]. The higher operating costs compared to greenhouses, due to the need for LED lighting and air conditioning, could be offset by reduced fuel costs to transport produce to market. The risk of losing crops due to occasional electricity outages can be less than the risk of losing crops in transport from rural areas to urban markets due to fuel shortages or absence of adequate cold storage. Another reason why completely enclosed structures could viable is that parameters can be set to provide optimum conditions year-round, enabling the higher running costs to be offset by higher, and more consistent yields [16].

We identified only two values from one specific study that provided estimates of energy consumption and GWP per kg [26] (Table 2). In this case, more energy (0.01/0.46 MJ/kg) are required during the dry season for irrigation than the seasonal systems for cold greenhouse and open field vegetables, respectively. The difference in the values of energy consumption is much less for the cold greenhouse than the open field because Beninese have no access to electricity for irrigation and use generators fueled with oil with a higher GWP [26]. This is in contrast to [11] who found that both energy and GWP were higher for heated greenhouses. The higher mineral and organic nitrogen fertilizer rate as well as irrigation efficiency were reported to have contributed to the difference of GWP due to both production of fertilizers and field emissions of 0.37 CO₂eq/kg in cold greenhouse and open field (0.11 kg CO₂eq/kg) vegetable production. The more water supplied, the higher the leaching rate and soil moisture content. The higher the maximal soil moisture, the higher the denitrification rate. The more nitrogen supplied, the more Nr emitted and soil pH that will increase the rate of volatilization. Overall, the nature and amount of energy consumed per volume of irrigation water applied were critical to the climate change potential.

Parameters	Greenhouse	Open field
Yield (kg/ha)	119,808	41,582
CED; MJ/kg	0.01	0.46
GWP; kg CO2eq/kg	0.37	0.11

Table 2.

Yield, cumulative energy demand and global warming potential for tomatoes, greenhouse and open field.

Source: Perrin et al. [26].

The maintenance of water pumps would limit the quantities of energy consumed as well as the irrigation efficiency. Second, it could also enhance crop yields at the edge of rivers where soils present a greater water retention capacity, lowering the need for irrigation water. Better irrigation management taking soil properties and local climate (evapo-transpiration) into account could improve the water use efficiency and also reduce (water losses by drainage).

Water use per kg product was based on ranges of water use per m² per day for tomatoes [26] and multiple vegetables [15, 25] (Table 3). These estimates are for vegetable production in Ghana where water is conveyed in 15-liter watering cans to irrigate. The range of values is large, 153–840 lts/kg, but it is consistent with estimates for field-based lettuce production in the USA of 201 lts/kg [11]. Both values are considerably larger than the 21 lts/kg reported by [11] for CEA leaf lettuce. The GWP of urban garden tomatoes in Benin were reported to be 4–23 times larger than the impacts of tomatoes grown in European cropping systems, due to low and variable crop yields (high fuel consumption for irrigation, large nutrient flows and use of insecticides [26]).

Parameters	Open field
	Maximum water usage	Minimum water usage
Yield, kg/ha	29,440	20,000
Yield, kg/m^2	2.944	2.000
Water, lts/m^2/day	5.0	14.0
Growing days	90	120
Water, lts/m^2	450	1680
Water, lts/kg	152.9	840.0

Table 3.

Estimated water use for open field [tomatoes and multiple vegetables].

Sources: Perrin et al. [26]; Obuobie et al. [15]; Drechsel and Keraita [25].

Given the very limited information on costs in most reviewed studies (e.g., [14, 15, 16, 18, 21]), improvements in data collection and reporting would be helpful to improve our understanding of the potential for CEA compared to field-based production systems. First, it is relevant to collect and report data for specific crops to facilitate comparison between production systems. CEA operations often focus on one or a few crops, so crop-specific data (i.e., not ‘vegetables’) is needed for adequate comparisons. Reporting of both yield per crop and yield per year when those are different would better represent total production for the purposes of calculating costs, revenue and input requirements per unit of product. CEA operations typically produce multiple crops per year, but this can also be true for field products (e.g., cabbage entries that have multiple cropping periods per year). Reasonably accurate cost data are also needed to make relevant comparisons. Future studies can usefully distinguish better between costs (both variable and fixed) and revenues. For example, some studies (e.g., [15, 16, 21]) report only price information (which can be used to calculate revenues) or total revenue information, but not cost data. Most of the studies reviewed reported no specific costs, either in aggregate (for a ha or for a cropping season) or per unit. It would also be helpful if additional disaggregation of cost categories (especially for energy inputs like fuel and electricity) were reported because they would facilitate improved estimates of environmental impact. In general, it would be helpful if future studies were also more comprehensive, reporting information on all performance metrics we considered: yields, costs, and input use for energy and water.

More studies of CEA are needed for Africa, especially for sub-Saharan Africa. This can include both less technologically advanced systems (e.g., greenhouses without full temperature or humidity control), and more advanced (and expensive) systems such as greenhouses with more environmental controls—and similar systems such as plant factories and vertical farms. Generally, greenhouses and polytunnel structures were readily obtained locally [16], except in Nigeria, where greenhouses are not yet popular (they are imported and relatively expensive [6, 16, 20, 21, 22, 23, 25, 26, 27, 28]). Variations the combine different characteristics may be relevant. For example, hydroponic units have been installed outside of greenhouses in Nigeria, influenced by crop varieties and space constraints. In Kenya, suppliers offer hydroponic units that can be installed in a variety of enclosed or open settings, including a small unit that can be mounted on the wall of a building for those with no land.

4. Conclusions

The uptake of CEA production technology has been limited in low and middle-income countries, particularly Africa. This means that we have very limited information to evaluate the potential of CEA and to make comparisons to field production. That is, until we have more examples—and data—to evaluate it will be difficult to understand the potential role for, and impacts of, different types of CEA production systems in the region. One requirement is expanded and improved data collection from existing operations. Improved data collection would include a broad range of relevant indicators collected in a consistent manner across farms and studies. Another approach is to develop more ‘synthetic’ approaches based on economic engineering approaches used for other food production technologies [28]. This approach can suggest the conditions under which CEA operations may be successful even in locations where they do not currently exist. The long-term success and economic viability of CEA in Africa will also depend on future trends in consumer preferences and market demand for vegetables (e.g., [25, 28]), so studies on the costs would be complemented by consumer preference studies. Together, this information will inform decisions about private investors and governments.

Conflict of interest

The authors declare no conflict of interest.

Authors contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by [Taiwo Ayinde], [Charles Fredrick Nicholson] and [Benjamin Ahmed]. The first draft of the manuscript was written by [Taiwo Ayinde] and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Ethical approval

All due consents have been sought.

Consent to participate

All due consents have been sought.

Consent to publish

All due consents have been sought.