Trends and Determinants of Virtual Water Trade and Water Resource Utilization in Ghanaian Vegetable Production

Abstract

Water plays a critical role in ensuring sustainable food security, particularly in the face of increasing freshwater scarcity and climate variability. This study examines virtual water use and virtual water trade in Ghana’s vegetable production sector over a 30-year period (1994–2023), focusing on four key crops: tomato, pepper, onion, and eggplant. Using secondary data on production volumes, trade flows, and virtual water content, the research quantifies imported and exported virtual water volumes and assesses net virtual water trends. The results reveal a substantial increase in virtual water use for most crops, with the exception of pepper, which experienced a marked decline. Onion and tomato are identified as the dominant contributors to both imports and exports of virtual water, while pepper and eggplant play relatively minor roles. The study finds that Ghana is a net importer of virtual water in vegetable trade, emphasizing the need for integrated water resource management to balance agricultural growth with water sustainability. A gravity model analysis was applied to identify the primary determinants of virtual water trade, revealing that GDP per capita, population size, distance, land availability, virtual water use, and border-sharing significantly influence trade patterns. The findings suggest that enhancing domestic production capacity and promoting efficient water use practices can reduce Ghana’s reliance on imports and improve resilience against water-related risks. This research provides valuable insights for policymakers, researchers, and practitioners aiming to develop sustainable water and food systems in Ghana and similar contexts.

1. Introduction

The persistent issue of water scarcity and its implications for food security has gained significant attention in the global context of water resource management and allocation strategies. Agriculture, accounting for more than 70 percent of global freshwater withdrawals, occupies a central position in addressing this key concern [1]. The sustainable goal of water resources management is critical for both environmental stability and economic productivity [2,3]. Water conservation and food security strategies enable national economies and agricultural sectors to effectively respond to the impacts of climate change while supporting the expanding population [4,5,6]. This aligns with the UN’s 2030 Sustainable Development Goals 1, 2, and 8, emphasizing the need for sustainable solutions to address global water and food security challenges [7].

Ghana’s agricultural sector is a vital pillar of its national economy, with vegetable production gaining prominence in ensuring domestic food security and international trade relations [8]. The country’s agricultural sector continually faces increasing pressure to meet the growing demand for vegetables. This surge, driven by population growth, urbanization, and shifting dietary preferences, presents substantial opportunities and considerable challenges [9]. Sustainable water resource management is a critical issue [10], particularly given Ghana’s vulnerability to water scarcity [11]. Climate change, manifesting in erratic rainfall patterns and increased frequency of droughts, further exacerbates this issue, placing immense strain on the nation’s water resources and threatening the long-term viability of its vegetable production sector [12]. The pressing challenge of fostering sustainable agricultural growth in Ghana necessitates a thorough understanding of the interplay between water and agriculture, which is fundamental for sustainable development [13,14,15]. Implementing effective strategies enhances agricultural productivity, mitigates environmental degradation, and bolsters climate resilience, thereby promoting the long-term sustainability of Ghana’s vegetable production sector [16,17,18].

In recent years, the concept of virtual water has received increasing attention as a strategic approach to managing and conserving water resources. Virtual water is the volume of water required to produce the goods and services people consume [19,20]. Consumers play a critical role in maintaining sustainable levels of human water consumption by minimizing their water footprint (WF), which refers to the volume of virtual water embedded in the products they consume [21]. Studies have demonstrated that reducing WF in agricultural production is a dependable approach to addressing local water scarcity and promoting ecological balance [22,23,24]. Furthermore, virtual water trade, which involves the exchange of goods and services based on the water required for their production, enables countries to optimize water use efficiency and supports agricultural resilience on a global scale [25]. The rising global demand for water resources is increasingly met through international trade, particularly in agricultural produce [26,27]. As a result, when a country imports food, it is also indirectly importing the virtual water embedded in the production of that food [28]. Notably, countries situated in arid regions often rely on virtual water imports as a strategy to mitigate the challenges associated with limited domestic water resources [29].

The gravity model, a well-established framework for analyzing spatial interactions such as trade, transportation, and migration flows [30], was utilized in the study to examine the determinants of virtual water trade involving Ghana. This trade model, inspired by Newton’s Law of Gravitation, posits that trade flows between two countries are proportional to their economic sizes and inversely related to the distance between them. It considers the exporter’s supply capacity, the importer’s demand potential, and trade costs represented by distance [29]. Since Tinbergen’s (1962) foundational work, the model has been widely used to analyze bilateral trade relationships [31] and, more recently, to assess the environmental impacts of trade by incorporating key variables such as population, income, and institutional factors within a robust economic framework [32]. Increasing scholarly attention has been directed toward understanding the factors that influence virtual water exchanges between nations, with the gravity model frequently serving as a methodological foundation for such investigations [33]. Since virtual water is transferred through the exchange of agricultural and industrial products, its driving mechanisms are inherently linked to both water resource endowments and socio-economic conditions. This study specifically explores how virtual water trade flows correlate with a set of key variables, including GDP per capita, geographical distance, population size, water use, land per capita, and a geographical contiguity dummy variable (border-sharing) between Ghana and its trading partners. This analytical framework facilitates the identification of the primary determinants influencing the global distribution of virtual water. Previous studies have highlighted these factors as significant drivers in shaping food trade patterns [29,34].

A limited understanding of the direct and indirect water consumption associated with the production of various goods can impede both consumers and traders from adopting water-saving practices [35]. In the Ghanaian context, where vegetables constitute a critical component of dietary intake, it is particularly important to analyze the water use patterns associated with their cultivation and trade. The trade-offs between import-related virtual water and export-related virtual water are shaped by a multifaceted set of economic, environmental, technological, geographical, and policy variables. Understanding these interrelated factors is vital for informed decision-making in the domains of water resource management, agricultural planning, international trade, and climate adaptation [36,37]. A thorough comprehension of these drivers not only facilitates the development of sustainable economic and environmental policies but also contributes to long-term water security and food system resilience. Such insights are instrumental in optimizing resource allocation, advancing agricultural sustainability, and supporting evidence-based policymaking, thereby aiding global sustainability initiatives.

Numerous studies on virtual water have examined global patterns [38,39], as well as regional and national flows of food commodities, primarily within developed countries [40,41]. However, empirical evidence on virtual water flows in Sub-Saharan Africa remains limited, particularly in the case of Ghana. One of the few exceptions is Tette et al. [42], who utilized recent data to provide detailed information on Ghana’s cereal and water security situation. This study adopted their approach with a specific focus on vegetable water use and the present dynamics of virtual water trade in Ghana. This was accomplished by investigating the amount of water expended in producing these key vegetables, analyzing the virtual water exchanged through the trade of these crops, and identifying the key factors that determine crop water use and virtual water trade. A thorough understanding of the findings presented in this study will inform the development of sustainable water management and food security strategies.

2. Materials and Methods

2.1. Study Area

Ghana, located in West Africa, has geographic coordinates ranging from latitudes 4.5° N to 11.5° N and longitudes 3.5° W to 1.5° E (Figure 1) [43]. The country is divided into 16 administrative regions. It shares its northern border with Burkina Faso, its eastern boundary with Togo, its western border with Côte d’Ivoire, and its southern boundary with the Gulf of Guinea and the Atlantic Ocean. The country covers a total land area of 238,533 km². As at 2023, Ghana’s population was estimated at 33.79 M, with approximately 41% residing in rural areas, where the majority engage in agricultural activities [44].

Ghana’s economy significantly depends on agriculture, particularly smallholder farming, which contributes 80% of the nation’s total agricultural output [43]. Agricultural land occupies 56% of the country’s total land area. In 2023, the agricultural sector contributed 21.1% to Ghana’s total GDP, reflecting an increase from 19.5% in 2022 [44].

Ghana’s precipitation patterns are influenced by the West African Monsoon (WAM), driven by the temperature and energy gradient between the Sahara Desert and the Gulf of Guinea. Rainfall variability is governed by the Inter-Tropical Discontinuity (ITD), which shifts north and south, resulting in either uni-modal or bi-modal rainfall patterns. Northern Ghana experiences a uni-modal rainfall pattern, occurring from April to mid-October, with peak rainfall in August or September. Southern Ghana has a bi-modal rainfall pattern, with the first season from March to July (peaking in June) and the second season from September to mid-November (peaking in October). From December to February, the harmattan, a dry north-easterly desert wind, dominates, particularly affecting northern Ghana by reducing humidity, resulting in hotter days and cooler nights [45,46].

Water withdrawal in Ghana constitutes roughly 2% of the entire national renewable water resources. Per capita annual water availability declined from 2890 m³ in 2000 to 1760 m³ in 2020, with an average decrease of 56.5 m³ per year, though the decline may not be linear [47]. Additionally, urbanization in Ghana grew from 52.8% in 2013 to 59.2% in 2023 [48]. This suggests that although urban development can stimulate economic growth, it often reduces water availability for agriculture. As water resources are increasingly reallocated to meet the demands of expanding urban areas, agricultural sectors may face significant constraints, ultimately affecting food production and undermining rural livelihoods.

2.2. Data Sources

This study focuses on four key vegetable crops in Ghana—tomato, pepper, onion, and eggplant—which are vital in enhancing food security and livelihoods [49]. Data on the import and export quantities of these crops from 1994 to 2023 were obtained from the FAOSTAT database [50]. The country-specific virtual water content (VWC) values of the crops were sourced from the estimates provided by Hoekstra and Mekonnen (2010) [51]. Data on land per capita, GDP per capita, and population for each country were sourced from the World Bank database [44]. The geographical distance between Ghana and its trading partners was determined using an online distance calculator [52].

2.3. Determining VWC and VWU

The crop virtual water use (VWU) for a given year was estimated by multiplying the country-specific VWC of each crop by the quantity of the respective commodity (Q) produced by a country [53]. A comprehensive analysis of agricultural production and trade necessitates a bottom-up approach, in which the virtual water content of each product is combined with raw production and trade data [30]. This relationship is mathematically expressed in Equation (1).

$$\mathrm{V}\mathrm{W}\mathrm{U}=\mathrm{Q}\times \mathrm{V}\mathrm{W}\mathrm{C}$$

(1)

VWU represents the total volume of water required for the production of a specific crop, encompassing all growth stages and production processes. It integrates three components of water use: green water, which refers to rainwater consumed through evapotranspiration during crop growth; blue water, which includes surface and groundwater resources that can be withdrawn for irrigation; and grey water, which accounts for the volume of freshwater required to dilute pollutants generated during crop production to meet water quality standards. The VWU is country-specific and is expressed in cubic meters per year (m³/yr). The quantity of crop production (Q) is measured in tons, while VWC is expressed in cubic meters per ton per year (m³/ton/yr).

2.4. Estimating IVW, EVW, and NVWT

The concept of virtual water flows is based on the idea that the exchange of products entails the transfer of the water embedded in their production [54]. The quantification of Imported Virtual Water (IVW) and Exported Virtual Water (EVW) is determined by the VWC of individual commodities and the corresponding trade volumes [55]. This relationship is formally expressed through established Equations (2)–(4) in the prior literature [42,54].

$$\mathrm{I}\mathrm{V}\mathrm{W}={\mathrm{Q}}_{\mathrm{i}}\times {\mathrm{V}\mathrm{W}\mathrm{C}}_{\mathrm{i}}$$

(2)

$$\mathrm{E}\mathrm{V}\mathrm{W}={\mathrm{Q}}_{\mathrm{e}}\times {\mathrm{V}\mathrm{W}\mathrm{C}}_{\mathrm{e}}$$

(3)

$$\mathrm{N}\mathrm{V}\mathrm{W}=\mathrm{I}\mathrm{V}\mathrm{W}-\mathrm{E}\mathrm{V}\mathrm{W}$$

(4)

The Net Virtual Water (NVW) represents the difference between a country’s IVW and EVW, indicating whether it functions as a net importer or exporter of virtual water. NVW, IVW, and EVW are measured in m³/yr. The variables ${\mathrm{V}\mathrm{W}\mathrm{C}}_{\mathrm{i}}$ and ${\mathrm{V}\mathrm{W}\mathrm{C}}_{\mathrm{e}}$ are importers’ VWC and exporters’ VWC per crop, respectively, measured in m³/ton. ${\mathrm{Q}}_{\mathrm{e}}$ represents the annual export volume (tons), and ${\mathrm{Q}}_{\mathrm{i}}$ is the annual import volume (tons).

2.5. Factors Influencing Virtual Water Trade

This study examines six drivers that affect virtual water trade, including one dummy variable. The selected drivers include GDP per capita, geographical distance, population size, water use, land per capita, and a geographical contiguity dummy variable (border-sharing) between Ghana and its trading partners. Several researchers have studied various influencing factors and their impacts on virtual water trade as shown in

Table 1. Previous and current gravity model studies on driving factors of VWT.

Referenced Study	Spatial Scale	Study Period	Findings
Tamea et al. (2014) [30]	Global	1986–2010	GDP, geographical distance, and population of both importer and exporter countries were fundamental drivers of VWF
Fracasso (2014) [56]	Global	2006	Traditional trade factors, national water availability, and water resource pressure levels influence bilateral VWF
Fracasso (2016) [37]	Global	2004	Larger water endowment does not necessarily lead to larger exports of VW
Tuninetti et al. (2017) [57]	Global	1986–2011	Key drivers of VWT include distance, population, and fertilizer application
Duarte et al. (2019) [58]	Global	1965–2010	Economic, geographical, institutional, and environmental factors had a significant influence on VWT
Garcia & Mejia (2019) [59]	Subnational	2007	Geographical distance, GDP, agricultural land, and population had a significant impact on VWT
Shivaswamy et al. (2021) [60]	Global	1990–2017	Distance is the key factor influencing VWT, while arable land availability and water usage are crucial in determining virtual water flows
Xia et al. (2022) [61]	Global	2000–2019	GDP and exchange rate positively influenced VWT but are negatively affected by arable land, per capita water resources, population, and geographic distance
Tette et al. (2024) [42]	Global	1992–2021	Significant drivers on VWT include GDP per capita, water use, land per capita, and population
Current study	Global	1994–2023	-

.

The selection of these variables is grounded in both theoretical frameworks and empirical findings from previous studies on international trade and water footprint analysis. The factors were chosen based on their potential influence on the volume and direction of virtual water flows between Ghana and its trading partners. Each of these variables was hypothesized to have a distinct impact on the magnitude and direction of virtual water trade. Their inclusion in the model allows for a nuanced understanding of the structural and geographic dynamics underlying Ghana’s virtual water exchanges.

2.6. The Gravity Model of Trade

The standard gravity model equation for bilateral trade is expressed in Equation (5) and serves as the theoretical and empirical foundation for this study. Specifically, the gravity model is employed to investigate the determinants of virtual water trade flows between Ghana and its trading partners. Using Eviews v.12 software, the panel least squares regression method was employed, with vegetable data comprising a combined total of tomato, pepper, onion, and eggplant. Due to missing data, the following six partners were not included in the analysis: Equatorial Guinea, Gambia, Guinea-Bissau, Federated States of Micronesia, New Caledonia, and Belgium-Luxembourg.

$${\mathrm{V}\mathrm{T}}_{\mathrm{j}\mathrm{k}}=\mathrm{G}({\mathrm{G}\mathrm{D}\mathrm{P}}_{\mathrm{j}}^{{\mathsf{\alpha }}_1}\times {\mathrm{G}\mathrm{D}\mathrm{P}}_{\mathrm{k}}^{{\mathsf{\alpha }}_2})/{\mathrm{D}}_{\mathrm{j}\mathrm{k}}^{{\mathsf{\alpha }}_3},$$

(5)

where VT is the volume of international trade (export or import), G represents the gravitational constant, GDP_j and GDP_k signify the Gross Domestic Product (GDPs) of countries “j” and “k”, respectively, and D represents the geographical distance between the trading partners (j and k). The superscripts ${\mathsf{\alpha }}_1$ and ${\mathsf{\alpha }}_2$ represent the elasticity of GDP variables while ${\mathsf{\alpha }}_3$ represents distance elasticity.

The gravity model is typically transformed into a log-linear form for estimation and this leads to the derived regression Equation (6), allowing for the inclusion of additional variables to assess their impact on trade volume.

$$\ln {\mathrm{V}\mathrm{T}}_{\mathrm{j}\mathrm{k}}=\mathrm{G}+{\mathsf{\alpha }}_1\ln {\mathrm{G}\mathrm{D}\mathrm{P}}_{\mathrm{j}}+{\mathsf{\alpha }}_2\ln {\mathrm{G}\mathrm{D}\mathrm{P}}_{\mathrm{k}}-{\mathsf{\alpha }}_3\ln {\mathrm{D}}_{\mathrm{j}\mathrm{k}}+\mathsf{\beta }\ln {\mathrm{V}}_{\mathrm{j}\mathrm{k}}+{\mathsf{\epsilon }}_{\mathrm{j}\mathrm{k}},$$

(6)

where $\mathsf{\beta }$ and ${\mathsf{\epsilon }}_{\mathrm{j}\mathrm{k}}$ represent the elasticity of the included variable ${\mathrm{V}}_{\mathrm{j}\mathrm{k}}$ and error term, respectively. Several studies have shown that gravity laws explain virtual water exports based on the characteristics of destination countries and virtual water imports as a function of the attributes of source countries [30,60]. However, this study examines the characteristics of both partner and host countries in relation to IVW and EVW to test the validity of the law. As a result, Equation (7) is formulated in line with previous studies [30,60,62].

$$\begin{array}{ll}\ln \left({\mathrm{V}\mathrm{W}\mathrm{T}}_{\mathrm{G}\mathrm{P}}\right)=& {\mathsf{\beta }}_0+{\mathsf{\beta }}_1\ln {\mathrm{G}\mathrm{D}\mathrm{P}}_{\mathrm{G}}+{\mathsf{\beta }}_2\ln {\mathrm{G}\mathrm{D}\mathrm{P}}_{\mathrm{P}}+{\mathsf{\beta }}_3\ln {\mathrm{L}}_{\mathrm{G}}+{\mathsf{\beta }}_4\ln {\mathrm{L}}_{\mathrm{P}}+{\mathsf{\beta }}_5\ln {\mathrm{P}}_{\mathrm{G}}+\\ & {\mathsf{\beta }}_6\ln {\mathrm{P}}_{\mathrm{P}}+{\mathsf{\beta }}_7\ln {\mathrm{V}\mathrm{W}\mathrm{U}}_{\mathrm{G}}+{\mathsf{\beta }}_8\ln {\mathrm{V}\mathrm{W}\mathrm{U}}_{\mathrm{P}}+{\mathsf{\beta }}_9\ln {\mathrm{D}}_{\mathrm{G}\mathrm{P}}+{\mathsf{\beta }}_{10}\ln \mathrm{B}+\mathsf{\epsilon },\end{array}$$

(7)

where variables with subscript G denote the host country, Ghana; while those with subscript P pertain to partner countries; VWT denotes virtual water trade representing either IVW or EVW over a 30-year period. L, GDP, D, P, VWU, and B refer to land per capita, GDP per capita, geographical distance, population, virtual water use, and geographical contiguity variable (shared-border), respectively; whiles ε represents their error term and β₁, β₂, …, β₁₀ are the parameter estimates.

3. Results

3.1. Virtual Water Use (VWU) Trends of Major Ghanaian Vegetable Crops

Figure 2 shows the changing trends in local production water footprint (virtual water use) for tomato, pepper, okra, and eggplant from 1994 to 2023. Tomato initially showed stable water use before experiencing a marked increase and maintaining a higher level in later years. In contrast, pepper experienced a sharp rise followed by a significant decline and eventual stabilization at lower levels. Onion and eggplant began with relatively low water use but demonstrated a consistent upward trend, particularly from 2007 onward.

3.2. Trends of IVW, EVW, and NVW

Figure 3 illustrates the IVW trends for tomato, pepper, onion, and eggplant over the period, revealing a significant shift beginning around 2006. Onion dominates import volumes, with recurrent peaks in 2012, 2016, and 2019, indicating growing reliance on external sources. Tomato also shows consistent, though moderate import activity with peaks in 2007, 2011, and 2018. In contrast, pepper and eggplant contribute negligibly to IVW throughout the period, suggesting domestic self-sufficiency.

As shown in Figure 4, EVW volumes for tomato, pepper, onion, and eggplant over a 30-year period indicate distinct crop-specific and temporal patterns. Onion exhibits the most substantial and episodic peaks, particularly around 2003 and 2007, indicating periods of a high export intensity. Tomato shows moderate and periodic export activity, with peaks in the early 2000s and around 2020. Eggplant contributes consistently but at lower volumes, while pepper displays minimal export activity with only occasional minor spikes. Overall, onion dominates in terms of export volume, followed by tomato and eggplant, with pepper playing a marginal role.

Figure 5 displays the net virtual water trade trends for four major vegetable crops over the study period, revealing significant interannual variability, particularly for onion. Onion exhibits the most pronounced shifts, with alternating periods of net imports and exports, and major export peaks between 2012 and 2019. Tomato shows moderate and relatively balanced fluctuations, while pepper and eggplant remain near zero throughout, indicating limited impact on net virtual water trade for Ghana. Overall, the volume IVW for Ghana across the combined crops exceeds that of EVW, indicating that Ghana is a net importer of vegetable virtual water.

3.3. The Gravity Model Analysis of VWT Drivers

The gravity model was utilized to examine 2250 cases over a 30-year period, covering 75 of Ghana’s 81 vegetable trade partners (

Table 2. Gravity model results of virtual water trade determinants of selected vegetables in Ghana.

Variable	IVW	EVW
C	89.945 (0.2058)	−106.707 (0.1955)
ln GDPG	−0.411 (0.8505)	−0.898 (0.4746)
ln LG	−3.27 (0.4038)	11.364 (0.0255)
ln PG	−4.136 (0.2821)	4.5764 (0.1421)
ln VWUG	1.571 (0.3935)	−2.3335 (0.0831)
ln DGP	0.007 (0.9873)	−0.932 (0.0472)
ln B	1.51 (0.1575)	2.024 (0.0299)
ln GDPP	−1.226 (0.0000)	0.157 (0.5082)
ln LP	−0.393 (0.0323)	0.031 (0.8713)
ln PP	−0.3766 (0.1299)	0.979 (0.0000)
ln VWUP	0.299 (0.027)	−0.38 (0.0052)
Adjusted R2	0.384	0.404
F-stat	15.051	10.703
Prob (F-statistic)	0.000	0.000

Note: p-values are reported in parentheses next to their respective regression coefficients.

and Figure A1). The regression analyses for Ghana’s virtual water imports (IVW) and exports (EVW) highlight notable differences in the significance and direction of key determinants. For imports, partner countries’ GDP per capita (GDP_P) and land per capita (L_P) show significant negative associations, suggesting that Ghana imports more virtual water from less developed countries with limited land resources. In contrast, partner countries’ vegetable virtual water use (VWU_P) positively influences imports, indicating stronger trade with water-intensive nations. However, domestic factors such as Ghana’s GDP per capita (GDP_G), land per capita (L_G), and population growth (P_G) do not significantly affect import flows. Geographical variables—distance (D_GP) and border (B) also appear statistically insignificant in this model.

For exports, the model reveals a different pattern. Ghana’s land per capita (L_G) significantly and positively impacts exports, underscoring the role of domestic resource endowments in supporting outbound trade. Partner countries’ population size (P_P) is a strong positive predictor, reflecting a higher demand from more populous nations. Notably, exports decrease with greater geographical distance (D_GP) and are higher for neighboring countries (B), indicating the importance of proximity in facilitating virtual water exports. Partner countries’ vegetable virtual water use (VWU_P) negatively affects exports, possibly reflecting reduced import needs in more self-sufficient nations. Both models are statistically significant overall, with adjusted R² values of 38.4% for imports and 40.4% for exports, indicating a moderate level of explanatory power in capturing the variation in virtual water trade flows of Ghanaians’ major vegetables.

4. Discussion

4.1. Virtual Water Use of Ghanaians’ Major Vegetable Crops

This study examined temporal trends in the virtual water use (VWU) associated with the local production of key vegetables—tomato, pepper, onion, and eggplant—in Ghana from 1994 to 2023. The findings reveal dynamic and crop-specific shifts in VWU over time, shaped by agricultural policy, technological adoption, and climatic variability. Notably, tomato exhibited a substantial and sustained increase in VWU after 2008, coinciding with national agricultural reforms and the expansion of irrigated horticulture. Conversely, pepper experienced a sharp rise followed by a decline and stabilization, reflecting possible shifts in production practices, pest pressures, and market demand [63]. Onion and eggplant demonstrated consistent upward trends in VWU, particularly from 2007 onward, likely due to growing consumer demand, improved irrigation access, and diversification of cropping systems [63]. These results align with the broader regional trajectory of agricultural intensification and increasing reliance on water resources to sustain food production [64].

Hoekstra and Mekonnen (2010) [51] investigated variations in virtual water content, illustrating the potential for substituting water-intensive crops with more sustainable alternatives to enhance water resource efficiency. Similarly, some studies have also examined economic factors influencing crop substitution, highlighting how farmers prioritize higher-value crops to maximize profitability [42,65]. This suggests that water availability and market-driven incentives are crucial in shaping agricultural production decisions, reflecting the critical importance of enhancing water resource management to meet increasing agricultural water demands exacerbated by climate change [66]. Given these dynamics, Ghana may consider prioritizing the expansion of local vegetable production, particularly tomato and onion, due to their high demand, significant market value, and relatively low VWC values. Presently, Ghana imports approximately 70% of its onion demand, incurring an estimated cost of USD 2 M weekly [67]. Similarly, the annual demand for fresh tomato is around 2.7 M metric tons [68]; however, in 2023, domestic production met only about 18.8% of this demand [50]. Increasing domestic production could reduce reliance on imports and create greater market opportunities for local farmers, thereby improving their livelihoods and saving foreign exchange for the country.

4.2. Ghana’s Vegetable Export, Import, and NVW

The exploration of virtual water imports and exports for tomato, pepper, onion, and eggplant highlights intricate trends and implications for sustainable water management, economic planning, and food security. The analysis of VWT over the 30-year period reveals a marked down trend in both IVW and EVW volumes. The combined decrease indicates that Ghana is increasingly self-sufficient in producing vegetables but progressively dependent on domestic water systems. This makes integrated management of water resources more crucial, especially under climate variability and increased population and for sustainable intensification of agriculture [42,54,69]. Considering Ghana’s growing reliance on domestic water resources for vegetable production, integrated water resource management (IWRM) is essential for sustainable agricultural development. Key strategies include promoting efficient irrigation technologies such as drip systems, adopting catchment-based planning, and enhancing inter-institutional coordination. Strengthening water monitoring through remote sensing and data systems, encouraging water reuse, and integrating climate adaptation into planning are also critical. Additionally, applying a water–energy–food nexus approach and empowering local water user associations will support equitable and resilient water governance. These measures collectively aim to optimize water use amid population growth, agricultural intensification, and climatic variability.

Despite the rise in water use driven by the expansion of vegetable production and the increasing effect of climate variability, Ghana remains a net importer of virtual water. This dependency on virtual water imports contributes to conserving domestic water resources while offering a potential opportunity to strengthen national food security. Most West African countries, including Ghana, more and more rely on imports of crops like onion and pepper during periods of low local production to meet their demand [19]. With Ghana’s abundant water resources [47], the country is strategically positioned to advance sustainable water management by expanding cultivation areas, enhancing climate resilience, improving irrigation infrastructure, and using high-yielding varieties to improve agricultural production capacity. For instance, in 2023, Ghana exported only 0.3% of its total eggplant production [50]. By leveraging these improvements, Ghana could increase the export of eggplant—water-intensive crop with the highest VWC among the selected crops—to water-scarce countries. This strategy would not only optimize virtual water trade but also support sustainable domestic water management.

4.3. Factors Influencing Ghana’s Vegetable VWT

Among the selected drivers, geographical distance, GDP per capita, population, land per capita, water use, and geographical contiguity dummy variable had a significant influence on VWT between Ghana and its partner countries. This reaffirms the core principle of the gravity model, indicating GDP and geographical distance as key determinants of trade [62]. However, the findings also diverge from earlier studies, which suggested that EVW is predominantly influenced by partner-country variables, while IVW is largely determined by host-country factors [30,60]. While the GDP of both host and partner countries plays a role in VWT dynamics, Ghana’s GDP does not exhibit a strong or clear effect on VWT flows in this study. Notably, several deviations from the existing literature emerge, including a negative association between partner-country GDP per capita and Ghana’s virtual water imports, the lack of statistical significance for geographical distance in the import model, and an unexpected negative relationship between Ghana’s own water use and its virtual water exports. These discrepancies may reflect Ghana-specific structural and institutional factors including trade policy orientations, the composition of agricultural exports, and infrastructural constraints within the agricultural sector. This study also suggests that geographical and biophysical variables, such as land and water availability, may be more prominent than purely economic indicators in explaining VWT flows, consistent with prior findings [70].

Additionally, the significant influence of land per capita and crop virtual water use aligns with the result of Shivaswamy (2021) [60], who emphasized the central role of land and water availability in agricultural trade flows. These findings also support Hoekstra and Chapagain (2008) [71], who noted that countries with greater land availability tend to be net virtual water exporters. Considering Ghana’s endowed water resources and extensive agricultural land [72], the nation holds significant potential to expand agricultural production, particularly vegetable cultivation for export, thereby reinforcing its comparative advantage in the regional and global markets.

Furthermore, the negative correlation between distance and virtual water flows confirms that greater distances reduce virtual water exports. This supports the traditional gravity model, where higher transportation costs and logistical constraints limit long-distance trade. In line with this, the significant effect of the geographical contiguity dummy variable indicates that neighboring countries receive more virtual water exports [2,30,73]. To enhance its virtual water trade flows, Ghana could leverage its regional trade dynamics by upgrading border facilities to streamline trade procedures and investing in better road networks, ports, and logistics infrastructure to lower transportation costs and minimize trade bottlenecks.

4.4. Limitations and Prospects

Although this study presents trends in virtual water use for the selected crops from 1994 to 2023, it utilizes fixed VWC values derived from a study by Mekonnen and Hoekstra (2010) [51], consistent with the approach adopted in previous research [30,42,74,75]. However, relying on these static values may not fully reflect present-day conditions. Factors such as climate variability, advancements in crop varieties, and improvements in agricultural practices over time are likely to have altered actual crop water use. This introduces a potential limitation to the study’s findings. To enhance the accuracy of future analyses, researchers are encouraged to adopt dynamically derived VWC estimates by incorporating crop water requirement calculations and actual yield data, as recommended by Chapagain and Hoekstra (2004) [39].

5. Conclusions

This study provides a comprehensive assessment of VWU and trade associated with key vegetable crops in Ghana—namely tomato, pepper, onion, and eggplant—over a 30-year period. The findings reveal distinct crop-specific trends in water use and trade patterns, influenced by shifting production priorities, climate variability, and evolving market demands. Notably, while the VWU for most vegetables has increased, pepper exhibited a decline, likely reflecting a reduction in production area or output. Onion and tomato emerged as the most significant contributors to Ghana’s virtual water trade, with onion particularly dominant in both imports and exports.

Ghana remains a net importer of virtual water in vegetable trade, which reflects its reliance on external sources to meet domestic demand during periods of limited local production. This dependency, while aiding domestic water conservation, also underscores the urgency for integrated water resource management (IWRM). Strategies such as efficient irrigation, climate-resilient infrastructure, and expanded domestic production—particularly of onion and tomato—can reduce this dependency and enhance national food and water security.

The gravity model analysis identified key determinants of Ghana’s virtual water trade flows. Partner countries’ GDP per capita, population size, and virtual water use, alongside factors such as land availability and geographical proximity, significantly influenced trade dynamics. These findings affirm the relevance of economic, geographic, and biophysical variables in shaping virtual water exchanges. Interestingly, certain results diverged from the established literature, suggesting Ghana-specific trade policies and institutional factors may be influencing VWT patterns.

In light of these insights, Ghana is well-positioned to capitalize on its agricultural potential by optimizing virtual water trade and enhancing water-use efficiency in vegetable production. A shift towards increased domestic output, supported by policy incentives and infrastructure investment, could not only mitigate import reliance but also bolster export capacity, particularly to neighboring water-scarce countries. Future research should incorporate dynamic VWC values and localized water-use data to better reflect contemporary production realities and further refine policy recommendations.