Sustainable Land Management by Agrivoltaics in Colombia’s Post-Conflict Regions: An Integrated Approach from the Water–Energy–Food Nexus

Abstract

Agrivoltaic (AV) systems are increasingly recognized as a strategy to enhance sustainable land management, yet their application in post-conflict settings remains underexplored. This study addresses this gap by evaluating AV deployment in two Colombian municipalities located in PDET/ZOMAC regions, using an integrated framework that expands the conventional Water–Energy–Food (WEF) nexus into the Water–Energy–Food–Soil–Climate–Communities (WEFSCC) nexus. The research combined GIS-based site characterization, crop yield and water balance modeling (contrasting traditional irrigation with hydroponics), and photovoltaic performance simulations for 30 kW systems, under conservative and moderate scenarios. Economic analyses included Net Present Value (NPV), Internal Rate of Return (IRR), and Free Cash Flow (FCL), with sensitivity tests for crop prices, yields, tariffs, and costs. Results indicate that AV can reduce crop irrigation demand by up to 40%, while generating 17 MWh/month of electricity per site. Cabrera exhibited higher profitability than Pisba, explained by yield differences and site-specific energy outputs. Comparative analysis confirmed consistency with experiences in Africa and Europe, while emphasizing local socio-environmental benefits. Conclusions highlight AV systems as resilient tools for sustainable land management in Colombia’s post-conflict regions, with actionable implications for land-use regulation, fiscal incentives, and international cooperation programs targeting rural development.

1. Introduction

The increasing global demand for food, water, and energy, coupled with land degradation and the impacts of climate change, has intensified the search for integrated approaches that maximize resource efficiency while minimizing environmental trade-offs [1,2,3]. Over the past four decades, agrivoltaic (AV) systems combining agricultural production with photovoltaic (PV) electricity generation on the same land have emerged as a promising solution to this challenge. First conceptualized by Goetzberger and Zastrow in 1982 [4], agrivoltaics were proposed as a response to the growing tension between energy production and agriculture, anticipating future land-use competition. Since then, the concept has evolved from theoretical models to large-scale pilot projects and commercial facilities, with more than a thousand systems currently operating worldwide [5,6,7,8]. Evidence from Europe, Asia, and Africa consistently demonstrates that agrivoltaics can simultaneously provide food and energy security while optimizing land productivity [9,10,11,12]. Reference [13] reported that land productivity can increase by up to 73% when crops are cultivated under PV modules, a finding that spurred the expansion of field trials in France, Italy, and Germany, and was conceptually reinforced by Dupraz [5] through the land-use efficiency framework. In arid and semi-arid regions, AV systems have been shown to mitigate soil moisture loss, lower soil surface temperatures, and reduce evapotranspiration by 20–60%, thereby improving water-use efficiency [10,14,15,16,17]. These benefits are particularly significant under conditions of water scarcity and climate variability, where shading from PV structures creates favorable microclimates for crops [10,14]. Technological advances have played a crucial role in expanding the feasibility of AV systems. The introduction of bifacial modules, vertical PV arrays, and single-axis trackers has enabled greater adaptability to diverse agricultural contexts [18,19,20]. For example, [5] documented significant improvements in both crop yields and energy efficiency under dual-use configurations in temperate regions. Likewise, [17,21] showed that in arid regions of the United States, shading from PV panels reduced plant stress, maintained soil humidity, and increased photosynthetic activity compared to control plots without panels. These results underscore the role of AV systems as buffers against extreme climatic conditions [2,15].

Beyond biophysical impacts, agrivoltaics also contribute to economic viability, social acceptance, and climate resilience. Studies in Asia and Europe confirm that integrating PV into agricultural landscapes can improve farm incomes by diversifying revenue streams, lowering irrigation costs, and ensuring a stable electricity supply for on-site consumption or grid injection [6,22,23,24]. For example, [25] assessed AV tomato production in Spain and found competitive yields with reduced water consumption, underscoring the potential of AV for horticultural applications in water-limited environments. Likewise, studies in East Africa and South Africa demonstrate that AV can simultaneously support rural electrification and food security, reinforcing its potential as a tool for sustainable development in regions with significant infrastructure gaps [9,26,27]. At the same time, AV systems contribute to climate mitigation and adaptation by promoting renewable energy generation, thereby supporting decarbonization goals aligned with the Paris Agreement and the EU Green Deal [8,28,29]. Furthermore, microclimatic regulation under PV structures reduces crop vulnerability to heat stress and drought, positioning AV as a technology that enhances resilience to climate extremes [10,13,14]. From a land management perspective, agrivoltaics foster synergies between energy and food sectors, reducing competition for land resources, an essential advantage in densely populated or resource-constrained countries [5,22,30]. Recent systematic reviews and bibliometric analyses confirm the rapid growth of AV research, particularly since 2015 [6,7,31]. These studies identify three main research clusters: (i) optimization of technical configurations (e.g., panel height, spacing, inclination, and transmissive materials) [18,32,33], (ii) evaluation of agronomic performance across diverse crops [10,25], and (iii) assessment of socio-economic and policy frameworks supporting AV adoption [8,24]. However, while evidence from Europe, North America, and Asia is robust, studies in Latin America and Africa remain comparatively limited, despite their abundant solar resources and growing agricultural pressures [9,34].

The international trajectory of agrivoltaics demonstrates consistent technical, agronomic, and socio-economic benefits that position AV systems as leading candidates for sustainable land-use innovation [6,7,35]. They have been shown to reduce evapotranspiration, enhance water-use efficiency, stabilize microclimates, and simultaneously produce clean energy and food [10,13,15,17]. Technological improvements continue to expand system adaptability [18,19,20], while economic assessments confirm profitability across diverse contexts [22,23,24]. This global evidence forms the baseline from which this study departs. However, despite these advances, there remains a gap in applying agrivoltaics to complex socio-political environments, such as post-conflict territories, and in frameworks that expand beyond the traditional WEF nexus to include soil, climate, and communities [8,36,37,38,39].

As example, in Latin America, the application of AV systems remains limited and fragmented. While pioneering projects have been implemented in Mexico, Brazil, and Chile, most are small-scale experimental plots or feasibility studies, with few systematic assessments of technical, economic, or social outcomes [6,19,34,35]. In Mexico, experiments have explored shade-tolerant crops under PV arrays [6,35]; in Brazil, research has examined integration with agroecological practices and community-based energy cooperatives [19,40]; and in Chile, studies have highlighted synergies with viticulture, demonstrating potential water savings and resilience to extreme droughts [17,34,41]. In Colombia, the evidence is even scarcer. Institutions such as the Unidad de Planificación Rural Agropecuaria (UPRA), the Instituto de Hidrología, Meteorología y Estudios Ambientales (IDEAM), and the Departamento Administrativo Nacional de Estadística (DANE) have identified the urgent need for sustainable land management and rural electrification, particularly in post-conflict regions [42,43,44,45]. However, most available documents remain limited to policy guidelines, technical potential assessments, or energy transition strategies [34,38,39,46], and unlike Europe or the United States, Colombia lacks demonstration sites or field-scale evaluations [34,38,39]. This absence is particularly critical in PDET and ZOMAC municipalities, prioritized in the 2016 Peace Agreement as territories requiring targeted investments for sustainable development [38,39]. These regions face overlapping challenges: limited connectivity to the national grid, high dependence on rainfed agriculture, degraded soils, and persistent socio-economic inequalities [45,46,47]. Although rural development policies emphasize renewable energy deployment and agricultural modernization, they have not yet considered agrivoltaics as a strategic tool to address these challenges simultaneously [43,44]. The lack of integration between energy transition and agricultural development agendas reinforces this gap, leaving post-conflict territories at risk of being excluded from innovative sustainability solutions [38,39,43,44,46].

A further research gap concerns the conceptual frameworks applied to agrivoltaics. Most international studies adopt the classical WEF nexus, focusing on resource efficiencies among water, energy, and food [1,3,30,48]. While this approach has proven useful, it remains insufficient to capture the complexity of fragile rural contexts such as those in Colombia. In these settings, soil degradation, climate variability, and community resilience are equally critical dimensions that determine the success of interventions [8,27,37,47,49,50]. Their omission from most AV research limits the capacity to fully assess long-term sustainability and socio-political feasibility. For instance, while water use efficiency and energy generation are central indicators, they fail to capture issues such as soil conservation, vulnerability to climate extremes, or the role of communities in governance and adoption [8,37]. Consequently, a clear twofold gap persists: first, the absence of empirical studies on agrivoltaics in Colombia and Latin America, particularly in post-conflict regions [19,34,38,39,41]; and second, the lack of analytical frameworks that integrate soil, climate, and communities alongside water, energy, and food [8,37,47]. This dual gap constrains the ability of policymakers and practitioners to design evidence-based strategies aligned with the Sustainable Development Goals (SDGs), the Colombian Peace Agreement, and international climate commitments [3,28].

Addressing these gaps requires moving beyond fragmented studies and adopting integrative frameworks capable of capturing the multidimensional role of agrivoltaics. Expanding from WEF to WEFSCC is not merely a conceptual innovation but a necessity to reflect the real challenges of Colombian post-conflict territories. By explicitly incorporating soil (land-use optimization and conservation), climate (variability and adaptation), and communities (social recovery and governance), the WEFSCC framework provides a more accurate lens for evaluating agrivoltaics in fragile contexts [27,47,49,50]. This perspective aligns with recent calls in the nexus literature to move from narrow technical assessments toward more holistic and policy-relevant analyses [1,2,3]. While agrivoltaics have demonstrated significant benefits internationally, few studies connect these systems to post-conflict sustainable development strategies, particularly in Colombia [34,38,39,46]. To address this gap, the present study integrates agronomic, hydrological, energy, financial, and socio-political dimensions within a WEFSCC framework [8,37,47,49], directly responding to the limitations identified in previous research [1,6,7]. Specifically, the research develops an integrative assessment of agrivoltaic systems as tools for sustainable land management in Colombia’s post-conflict regions, focusing on two municipalities—Pisba (Boyacá) and Cabrera (Cundinamarca)—both classified as PDET/ZOMAC territories [43,44,45,46]. These areas represent national priorities for sustainable development but remain constrained by infrastructure deficits, land-use conflicts, and high vulnerability to climate variability.

The main contribution of this work lies in bridging the technical, economic, and socio-environmental dimensions of agrivoltaics within a single analytical structure that combines quantitative modeling with contextual interpretation [6,7,8]. Specifically, the research integrates: (i) agroclimatic and spatial analysis, using GIS v3.32 tools and IDEAM datasets were obtained using the official web platform Consulta y Descarga de Datos Hidrometeorológicos—DHIME to characterize solar irradiance, slope, and land-use conditions [45,51,52]; (ii) crop yield and water balance modeling, contrasting traditional surface irrigation with hydroponic systems to estimate reductions in evapotranspiration and water demand [36,53,54]; (iii) photovoltaic energy simulation, evaluating the generation potential of 30 kW systems under site-specific conditions [19,20,33]; (iv) economic evaluation, incorporating Net Present Value (NPV), Internal Rate of Return (IRR), Free Cash Flow (FCL), and Levelized Cost of Energy (LCOE) under different scenarios [22,23,24]; and (v) socio-environmental interpretation, examining how agrivoltaics contribute to community resilience, land restoration, and climate adaptation within post-conflict reconstruction programs [27,38,39,47,50]. By integrating these components, the study addresses the fragmentation that characterizes most previous research, which often isolates either the technical or the socio-economic aspects of AV systems [1,6,7]. This multidimensional approach not only quantifies potential gains in water, energy, and agricultural productivity but also identifies synergies and trade-offs across the six dimensions of the WEFSCC nexus [5,22,28]. For example, while hydroponic AV systems improve water efficiency and crop yields, they also require higher initial investments and technical capacity, which may influence community adoption and equity outcomes [24,25,27,55].

Another major contribution of this study is the contextualization of agrivoltaics within Colombia’s peacebuilding agenda. Post-conflict regions in Colombia represent unique socio-ecological systems where land-use planning intersects with social reconciliation and climate adaptation [27,43,44,47]. By situating agrivoltaic deployment in PDET and ZOMAC municipalities, the study demonstrates how renewable energy solutions can directly contribute to Sustainable Development Goals (SDGs) 2 (Zero Hunger), 7 (Affordable and Clean Energy), 13 (Climate Action), and 16 (Peace, Justice, and Strong Institutions) [3,8,28]. This alignment also provides a replicable model for other Global South countries facing similar challenges of rural inequality, post-conflict reconstruction, and ecological vulnerability [50,56,57]. Furthermore, the study contributes methodologically by offering a scalable decision-support framework. The integration of spatial, hydrological, and financial modeling allows replication in other regions [22,23,36], while the inclusion of the WEFSCC perspective ensures that results can inform both engineering design and policy formulation [8,37,47]. This approach aligns with the current shift in nexus research toward actionable, transdisciplinary applications that connect scientific evidence with governance and planning processes [1,2,3]. In addition, this paper contributes to the broader scientific debate by operationalizing the WEFSCC nexus through empirical data and scenario analysis [8,37]. While the WEF nexus has provided a valuable foundation [1,3,48], its expansion to include soil, climate, and communities allows a more comprehensive understanding of system interdependencies. This conceptual advance responds directly to recent calls for a “WEF-plus” paradigm capable of incorporating equity, resilience, and environmental feedback into decision-making [2,27,50]. By doing so, this study not only provides a country-specific assessment for Colombia but also enriches global discussions on the role of agrivoltaics in sustainable land-use transitions. Importantly, this research advances the understanding of agrivoltaics as catalysts of territorial resilience. In Colombia, where historical conflicts have disrupted rural livelihoods and degraded natural resources, agrivoltaic systems can become platforms for social innovation—combining local participation, technology transfer, and environmental restoration. The findings suggest that agrivoltaics not only improve resource efficiency but also reframe how land is valued and used, transforming spaces of conflict into spaces of cooperation.

These interdependencies generate both synergies and trade-offs that must be explicitly modeled to inform sustainable planning [2,8,58]. From a policy perspective, adopting a WEFSCC approach enables governments to design interventions that transcend sectoral boundaries. In Colombia, applying this framework can help bridge the gap between the energy transition agenda and rural development policies, ensuring that renewable energy deployment simultaneously supports agricultural productivity, soil conservation, and peacebuilding [38,39,43,44,46]. The WEFSCC nexus thus provides a foundation for cross-sectoral coordination, facilitating the development of incentives and governance instruments that reflect the integrated nature of resource systems.

In summary, this conceptual advancement ensures that agrivoltaic systems are evaluated not only in terms of efficiency metrics but also in relation to ecological integrity, climate resilience, and social justice principles that underpin the sustainable reconstruction of rural Colombia. In light of these conceptual and empirical gaps, this study evaluates how agrivoltaic (AV) systems can function as integrative instruments for sustainable land management in Colombia’s post-conflict territories, framed within the Water–Energy–Food–Soil–Climate–Communities (WEFSCC) nexus. Addressing sustainability challenges in fragile regions requires not only technological innovation but also a systemic understanding of the socio-ecological interdependencies that shape these landscapes. Accordingly, the research adopts a transdisciplinary approach that links engineering analysis with environmental, economic, and social dimensions, aiming to generate actionable insights for both policy and practice. The overall objective of this work is to assess the technical, economic, environmental, and social feasibility of agrivoltaic systems in selected post-conflict municipalities of Colombia through an integrative WEFSCC framework that captures resource interdependencies and identifies pathways toward sustainable land management and community resilience. To achieve this, the study first characterizes the agroclimatic, hydrological, and spatial conditions of the municipalities of Pisba and Cabrera, both part of the PDET and ZOMAC programs, using QGIS-based tools v3.32 and national datasets (IDEAM, UPRA, DANE) to evaluate their suitability for agrivoltaic deployment. It then models crop productivity and water consumption under traditional irrigation and hydroponic systems, quantifying the potential reductions in evapotranspiration and water demand under agrivoltaic shading. Photovoltaic energy generation for 30 kW systems is simulated at both sites to estimate hourly, daily, and monthly energy outputs under varying tilt and solar incidence conditions. Economic performance is analyzed through Net Present Value (NPV), Internal Rate of Return (IRR), and Free Cash Flow (FCL) indicators, under conservative, moderate, and optimistic scenarios. Finally, all results are integrated within the WEFSCC framework to identify synergies and trade-offs among the six resource dimensions (water, energy, food, soil, climate, and communities) and to derive implications for local governance, policy design, and sustainable territorial planning. Building on these objectives, the study is guided by the following research questions: (i) How can agrivoltaic systems contribute to reducing water demand and improving crop productivity in rural post-conflict territories? (ii) What is the expected energy generation potential and economic viability of small-scale agrivoltaic installations (30 kW) under Colombian agroclimatic conditions? (iii) How do soil, climate, and community dynamics modify the traditional interactions of the WEF nexus in fragile territories? (iv) What policy instruments and governance mechanisms could support the adoption of agrivoltaic systems as a strategy for land restoration, climate adaptation, and peacebuilding? By articulating these objectives and research questions, this study aims to provide a comprehensive and replicable framework for assessing agrivoltaics within complex socio-environmental contexts. The proposed approach not only quantifies the technical and financial viability of dual-use systems but also situates them within a broader narrative of sustainability, resilience, and post-conflict reconstruction. This orientation ensures that the findings are relevant to engineers, planners, policymakers, community organizations, and international cooperation agencies committed to promoting integrated development pathways. The subsequent sections describe in detail the methodological structure, analytical models, and data sources employed to operationalize the WEFSCC nexus and evaluate agrivoltaic potential in Pisba and Cabrera.

2. Conceptual Framework: The WEFSCC Nexus

The conceptual framework guiding this study extends the conventional Water–Energy–Food (WEF) nexus to incorporate three additional and interdependent dimensions—Soil, Climate, and Communities, forming the WEFSCC nexus. This expansion responds to the growing recognition that while the classical WEF model is valuable for resource optimization, it is often insufficient to capture the complex socio-ecological realities of rural territories in the Global South, particularly in post-conflict regions such as those in Colombia. The WEF nexus traditionally emphasizes the quantitative interdependencies among resource flows: water availability affects food production; food systems depend on energy for irrigation and processing; and energy systems, in turn, require water for generation and cooling [1,3,48]. However, this triad often treats land, climate, and social systems as external boundary conditions rather than integral components of resource governance. Recent studies have criticized this limitation, stressing that the exclusion of soil quality, climate feedback, and community dynamics constrains both the explanatory and practical power of nexus models [27,49,50]. In response, broader frameworks—such as WEF-Land, WEF-Ecosystem, or WEF-Climate-Community nexuses—have been proposed to explicitly integrate ecological and social variables, thereby enhancing resilience, equity, and sustainability [8,37,51].

The Soil Dimension represents the foundational substrate linking food, water, and energy systems. In agrivoltaic configurations, soil is not merely a passive medium for crop growth; it also shapes hydrological processes, microclimate regulation, and land stability [47,49]. The installation of PV structures alters soil albedo, temperature, and moisture retention, which can either improve or degrade soil health depending on design and management [9,14]. Studies in arid zones demonstrate that partial shading reduces soil evaporation, increases microbial activity, and enhances carbon sequestration [13,14,16,35,47]. Conversely, poorly designed installations may cause compaction, erosion, or hydrophobicity. For Colombia, where land degradation affects more than 30% of agricultural soils, the soil dimension is central to sustainable land-management strategies [42,44,47]. Integrating soil into the nexus enables evaluation of how agrivoltaic systems contribute to conservation, erosion control, and the restoration of degraded lands, aligning with national goals under the Misión de Suelos and the UN Decade on Ecosystem Restoration.

The Climate Dimension positions climate as both a driver and a constraint in WEF interactions. Variability in precipitation, temperature, and solar radiation directly shapes agricultural productivity, irrigation demand, and PV performance. In the classical WEF framework, climate is usually treated as an exogenous factor; within WEFSCC, it becomes endogenous a feedback variable that both influences and is influenced by resource management decisions [1,2,3]. Agrivoltaics provide a distinctive pathway for climate adaptation and mitigation: by reducing evapotranspiration and creating cooler microclimates, they help crops withstand heat and drought stress [13,15,17,58], while simultaneously generating renewable energy that offsets greenhouse gas emissions from fossil sources. In this sense, AV systems operate as climate-smart infrastructures. Their dual capacity to mitigate and adapt makes them especially relevant in Colombia, where ENSO-driven variability causes recurrent droughts and floods that destabilize agricultural output [42,45,51]. Thus, the WEFSCC framework integrates climate not as a background condition but as a co-evolving dimension whose stability depends on sustainable land and energy practices.

The Communities Dimension constitutes the governance backbone of any nexus system. Resource flows are not managed by institutions alone but also by social networks, cooperatives, and local stakeholders, whose participation determines the long-term success of technological interventions [3,7,27]. Within WEFSCC, the community dimension emphasizes agency, equity, and inclusion elements largely absent from technocratic WEF models. In Colombia’s post-conflict territories, communities are rebuilding livelihoods amid historical inequities and institutional fragility [38,39,59]. Agrivoltaic projects in such contexts can foster social cohesion by providing shared energy infrastructure, generating employment, and strengthening food security, while also serving as platforms for education and demonstration [26,40,57]. Moreover, the inclusion of communities enables the nexus to capture feedback loops between governance and resource efficiency: empowered communities manage water and land more sustainably, while trust and participation reduce project failure rates [8,50,60]. In this sense, the WEFSCC framework aligns with a human-centered sustainability paradigm, linking technical outcomes with social well-being.

Taken together, the Soil, Climate, and Communities dimensions complete the WEFSCC lens through which we examine agrivoltaics in post-conflict territories; the next section operationalizes this framework for Pisba and Cabrera, detailing the datasets (IDEAM, UPRA, DANE), spatial and agroclimatic screening, crop–water balance modeling under surface irrigation and hydroponics, 30 kW PV generation simulations, and the economic evaluation (NPV, IRR, FCL, LCOE) and sensitivity analyses, as well as the integration procedure used to map WEFSCC synergies and trade-offs and derive governance-relevant insights.

3. Materials and Methods

3.1. Study Area and Site Selection

The study was conducted in two rural municipalities of Colombia, Pisba (Boyacá) and Cabrera (Cundinamarca), both classified as priority territories under the Development Programs with a Territorial Focus (PDET) and the Zones Most Affected by Armed Conflict (ZOMAC) frameworks established after the 2016 Peace Agreement. These municipalities represent contrasting agroecological and socioeconomic conditions typical of Colombia’s Andean region and were selected for their relevance to post-conflict rural development and their suitability for implementing agrivoltaic (AV) systems within the Water–Energy–Food–Soil–Climate–Communities (WEFSCC) nexus [7,8,9,44,48,49].

Pisba is located in the northeastern sector of Boyacá, within the high Andean zone of the Eastern Cordillera, at coordinates 5°43′21″ N, 72°29′07″ W, with altitudes ranging from 1300 to 3400 m.a.s.l., with average temperatures between 13 and 20 °C and annual precipitation exceeding 1200 mm. According to data from the Integrated Agricultural Planning System (SIPRA) of UPRA, Pisba has an agricultural frontier of approximately 12,246 ha, as illustrated in Figure 1.

Figure 1. Agricultural frontier and topographic configuration of Pisba (Boyacá, Colombia). The boundaries of the municipality and its agricultural frontier (12,246 ha) are shown according to SIPRA–UPRA datasets. Green areas indicate current vegetation and forest cover, while the purple outline indicates the administrative limits. The accompanying bar scale graphic illustrates the municipal agricultural frontier extension and is included to contextualize spatial magnitude.

Cabrera is located in the southern province of Sumapaz, Cundinamarca, at coordinates 3°58′41″ N, 74°29′09″ W, with altitudes between 1800 and 2400 m.a.s.l., with a mean annual temperature of 15–22 °C, annual precipitation exceeding 1000 mm and an agroclimatic setting classified as montane humid forest [50,51]. According to data from the Integrated Agricultural Planning System (SIPRA) of UPRA, Cabrera has an agricultural frontier of approximately 14,385 ha, as illustrated in Figure 2.

Figure 2. Agricultural frontier and topographic configuration of Cabrera (Cundinamarca, Colombia). The boundaries of the municipality and its agricultural frontier (14,385 ha) are shown according to SIPRA–UPRA datasets. Green areas indicate current vegetation and forest cover, while the purple outline indicates the administrative limits of Cabrera within the Sumapaz region. The accompanying bar scale graphic illustrates the municipal agricultural frontier extension and is included to contextualize spatial magnitude.

For both municipalities, geographic, climatic, and productive data were collected using GIS tools (QGIS 3.28). Data such as solar radiation, precipitation, temperature, evapotranspiration were obtained from the IDEAM meteorological stations located in each case study area: Pisba—[35215030] (Boyacá, Pisva), categorized as Climática Ordinaria, at an altitude of 1482 m.a.s.l. (latitude 5.72° N, longitude −72.49° W), with automatic telemetry and active since 15 December 2016; Cabrera—[21190090] (Cundinamarca, Cabrera), categorized as Pluviométrica, at an altitude of 1900 m.a.s.l. (latitude 3.99° N, longitude −74.48° W), with conventional instrumentation and active since 15 September 1958, but also with technical field visits [45,52].

Climatic series covered a 10-year (2010–2020) period for maximum and minimum temperature, wind speed, relative humidity, and solar radiation, while a 20-year series (2000–2020) was used for precipitation to ensure robust representation of interannual variability. The reference evapotranspiration (ET_o) was calculated using the FAO Penman–Monteith equation, and crop water requirements were derived through the ET_o–K_c workflow, applying crop coefficients (K_c) validated for leafy vegetables (lettuce, spinach) and strawberries under partial shade conditions.

The selection of Pisba and Cabrera was based on a multi-criteria evaluation that combined technical, environmental, and social parameters. First, both territories display high solar availability and agricultural vocation, with extensive rural areas lacking connection to the National Interconnected System (SIN). Second, their inclusion in the PDET/ZOMAC programs highlights the need for innovative technologies that promote local employment, sustainable resource management, and energy access. Third, the contrasting agroclimatic conditions of Pisba (humid, high-altitude) and Cabrera (sub-humid, mid-altitude) allow for comparative analysis of AV system performance under different microclimatic and topographic contexts. Finally, both municipalities possess institutional partnerships with local agricultural associations and regional environmental authorities, facilitating the integration of community participation within the WEFSCC framework. Although the analysis focused on Pisba and Cabrera, the selection criteria solar resource availability, altitude range (1800–3400 m.a.s.l.), and agricultural frontier characteristics are also met in other highland municipalities of Cundinamarca such as Tocancipá, Susa, and Nemocón. These areas exhibit analogous agroclimatic and topographic profiles, suggesting strong potential for replicating the agrivoltaic–hydroponic configuration under the same WEFSCC framework.

The characterization of these sites therefore captures the spatial heterogeneity of post-conflict territories and provides a robust foundation for modeling the technical, hydrological, and economic behavior of agrivoltaic systems. The geographic coordinates, long-term actinometric datasets, and agro-environmental attributes defined in this subsection serve as baseline inputs for the subsequent analyses of crop productivity, water consumption, and photovoltaic energy generation.

Potential plots, available electrical infrastructure, and edaphic constraints were identified with the support of UPRA/SIPRA and FAO criteria [43,47]. The availability of electrical infrastructure for local-scale energy integration was verified and contextualized within the national energy balance provided by XM [45].

3.2. Agrovoltaic Conceptual and Analytical Framework Design

The methodology is based on the design and implementation of an agrivoltaic system integrated with vertical hydroponics, referred to in this study as the Productive Unit (PU). This system is conceived to address simultaneously the three core pillars of the nexus: (i) Water, by reducing consumption through the use of hydroponic crops grown in vertical recirculating towers with controlled losses, complemented by rainwater harvesting. (ii) Energy, by generating clean electricity from photovoltaic solar panels to meet the system’s demand while producing surplus energy. (iii) Food, by increasing agricultural productivity per unit of land area through the cultivation of high-efficiency crops with low water requirements.

This integrated approach combines technical modeling, environmental assessment, and socio-economic interpretation, ensuring coherence between physical processes and governance outcomes. The methodological design was structured within the Water–Energy–Food–Soil–Climate–Communities (WEFSCC) nexus to evaluate agrivoltaic (AV) systems as holistic instruments for sustainable land management in post-conflict regions. This framework is supported by previous studies that demonstrate the effectiveness of AV systems in articulating resource flows under conditions of aridity and land-use pressure [13,16,35], as well as their capacity to enhance socio-ecological resilience in rural environments [58]. The overall methodological workflow from knowledge synthesis and conceptual design to subsystem modeling, integration, and system assessment, is summarized in Figure 3, which illustrates the sequence from data collection to the integration of results and the evaluation of system adaptability.

Figure 3. Methodological workflow under the WEFSCC nexus for evaluating agrivoltaic systems in post-conflict territories of Colombia.

The whole process begins with a comprehensive state of the art review, focused on technological, agronomic, and environmental components relevant to AV systems, including hydroponic configurations, solar photovoltaic technologies, and water-saving strategies. In parallel, a design requirements analysis was conducted to determine structural, operational, and environmental constraints under Colombian conditions. The design process of the agrivoltaic system was carried out through an iterative method focused on digital conceptualization and validation, with the objective of optimizing both energy efficiency and structural integration within the agricultural environment. The first step of the project design began with the generation of preliminary concepts that considered the arrangement of solar panels, support modulation, and interaction with the cultivated space. Each concept was modeled using 3D CAD software (Fusion 360 June 2024 release and Inventor 2024 with the FEA module),allowing for the analysis of proportions, assemblies, and potential interferences from the early stages [57]. This approach facilitated the exploration of multiple configurations prior to moving into more detailed simulations.

Subsequently, structural and performance simulations were conducted to evaluate the resistance of the supports, stability under environmental loads, and stress distribution at critical points, guiding successive adjustments in the geometry and ensuring an adequate safety factor prior to fabricating any physical prototype. This iterative process of modeling and simulation, validated by recent research on AV in different regions [9], allowed the design to be refined until achieving a digitally validated functional model, represented through renders and animations. In this way, the design ensured a balance between energy efficiency, structural reliability, and implementation feasibility. These inputs informed the formulation of the conceptual model and the selection of simulation tools for water, energy, and crop performance.

The validated conceptual design then informed the technical modeling stage, developed through three parallel subsystems: (i) agricultural and hydroponic performance, represented by crop yield estimation and evapotranspiration modeling; (ii) photovoltaic energy generation, simulated using hourly irradiance and temperature data; and (iii) hydrological balance, evaluating irrigation needs and water recovery under AV shading. Each subsystem was parameterized using climatic, edaphic, and geographic data from IDEAM, UPRA, and NASA-POWER databases. The resulting datasets were harmonized to ensure cross-compatibility and integrated into a common temporal resolution (hourly–monthly).

After subsystem integration, the combined outputs were analyzed under the WEFSCC nexus, incorporating quantitative and qualitative variables. Quantitative indicators included water-use efficiency, energy generation, and economic performance metrics such as Net Present Value (NPV), Internal Rate of Return (IRR), and Free Cash Flow (FCL); when pertinent to investment comparisons, CapEx and OpEx were also considered. Qualitative dimensions such as soil optimization, climate variability, and community participation were examined through semi-structured stakeholder interviews, document analysis, and expert consultation to capture governance factors and contextual constraints that condition adoption and scalability.

A feedback mechanism was incorporated to assess system adaptability. If results indicated that technical or environmental conditions limited scalability, the AV configuration was redesigned considering regional constraints, including soil conservation needs, local climatic variability, and community resource governance. Otherwise, the process advanced toward pilot implementation, performance testing, and long-term monitoring. This iterative structure ensures that the model remains dynamic, context-sensitive, and regionally adaptable.

Overall, this methodological framework enables a comprehensive understanding of agrivoltaic systems beyond their technical efficiency. It captures how water, energy, and food flows interact with soil, climate, and social dimensions, thereby operationalizing the WEFSCC perspective in a quantifiable manner. In addition, the analytical structure provides a foundation for scenario testing, allowing policymakers and practitioners to explore “what-if” conditions related to resource trade-offs and policy interventions and to inform sustainable land-use transitions in post-conflict territories.

Technical Components of the System

The photovoltaic system was designed with Risen Hyper-ion bifacial modules RSM132-8-705BHDG, with a rated capacity of 705 Wp and an efficiency of 22.7%. A total of 43 modules were installed per site, yielding a PV DC nameplate capacity of 30 kWpDC. Modules were mounted at a height of 2.5 m on modular structures oriented due south (azimuth 180°) with a fixed tilt angle of 10–15°, following technical recommendations for agrivoltaics in tropical latitudes but also following technical criteria reported in agrivoltaic projects in Europe and Asia that emphasize the importance of height and orientation for optimizing both energy and agricultural performance [5,10,55]. The installed capacity was the same for Pisba and Cabrera with 43 modules. Stringing configuration consisted of three strings (15S, 14S, 14S) connected to the inverter, ensuring maximum open-circuit voltage and operating voltage remain within equipment limits.

The selected inverter was a Huawei SUN2000-30KTL-M3, with a rated 30 kWAC output and four independent MPPT inputs, providing a DC/AC ratio of 1.0, which lies within recommended design ranges. The installed capacity per municipality was therefore approximately 30 kWpDC/30 kWAC.

The photovoltaic system losses were estimated by considering the main operating factors under real field conditions. Table 1 shows the breakdown of each loss component and the calculation of the Performance Ratio (PR), using the thermal coefficients of the selected module and conservative assumptions for operation and maintenance.

Table 1. Thermal Loss and PR with Module Coefficients.

Component	Loss	Comment
Temperature (with αPmax)	5%	With NMOT = 43 °C
Soiling	3%	Seasonal cleaning
Agrivoltaic shading	7%	Towers/structure; mitigated layout
Mismatch	2%	Module dispersion
DC resistive losses	2%	Voltage drop ≤1.5%
Inverter efficiency	3%	η Euro/CEC
Availability/O&M	1%	Scheduled
Total PR (product)	0.790

The support structures, built with guadua (a native bamboo), treated with weather-resistant resins, were designed in accordance with the Colombian Seismic Resistant Code NSR-10, considering wind and seismic loads. This approach aligns with studies that promote the use of sustainable local materials in AV infrastructure [55].

For the agricultural system, five-level hydroponic towers of the NFT type (Nutrient Film Technique) were employed, with a recirculating nutrient solution, submersible pump of 1.5 hp, and pH and EC controllers. Such configurations have been shown to increase water-use efficiency by up to 60% compared to traditional agriculture [41,61]. Each tower level contains 20 planting holes, reaching a cultivation density of approximately 100 plants/m². In total, 43 towers were installed in Pisba and Cabrera.

This design integrates clean energy technologies and soilless agriculture, generating a replicable scheme for rural contexts with water and land limitations, in line with international experiences in agrivoltaics combined with hydroponics [17,62].

3.3. Crop Selection and Yield Estimation

Crop selection in agrivoltaic–hydroponic systems represent a determining factor for the performance and adaptability of the productive model, since it influences water-use efficiency, light interception, physiological resilience under partial shading, and the economic viability of the integrated unit. In this study, the selection process was conducted using three primary criteria: (i) physiological compatibility with semi-shade conditions; (ii) adaptability to hydroponic systems with short growth cycles and high turnover; and (iii) profitability and local market demand in the selected municipalities.

Based on these parameters, three crops were selected: lettuce (Lactuca sativa), spinach (Spinacia oleracea), and strawberry (Fragaria × ananassa). These species are widely documented for their efficient water use, short phenological cycles, and commercial acceptance in Andean and inter-Andean regions. Lettuce and spinach are typical leafy vegetables well suited to hydroponic NFT or vertical recirculation systems, while strawberry represents a reference crop of higher market value but longer cycles. Previous studies report that these crops maintain stable yields under moderate shading (up to 30–40%) without compromising physiological integrity or quality [63,64].

3.3.1. Agroclimatic Compatibility and Site Conditions

The agroclimatic characterization for Pisba (1300 to 3400 m.a.s.l.) and Cabrera (1800 and 2400 m.a.s.l.) indicates average annual temperatures of 15–18 °C and mean radiation values between 300 and 500 W m⁻², according to IDEAM historical data. These parameters correspond to the tierra fría and tierra templada zones, suitable for leafy crops with moderate light requirements. The selected species exhibit optimal photosynthetic performance under these ranges, aligning with conditions described in the Manual de Hidroponía [65] and the Manual de Recomendaciones Técnicas para Cultivo de Fresa en Cundinamarca [66].

3.3.2. Hydroponic System and Growth Parameters

All crops were modeled under a vertical hydroponic system with recirculation, where nutrient-enriched water flows through stacked towers. Each module integrates an NFT-type circulation loop, minimizing water losses through evapotranspiration and leakage. The nutrient solution maintains a pH between 5.5 and 6.5 and electrical conductivity (EC) between 1.5 and 2.0 mS cm⁻¹, ensuring macro- and micronutrient balance [67]. The towers operate with a solution turnover of 3–4 L per plant per cycle [65].

The general physiological and management parameters used as the basis for simulation are summarized in Table 2, adapted from Refs. [65,67].

Table 2. Cultivation parameters for selected crops under hydroponic agrivoltaic conditions.

Crop	Optimal Temperature (°C)	Solar Radiation $(\mathbf{W}/{\mathbf{m}}^2)$	pH	EC $(\mathbf{m}\mathbf{S}/\mathbf{c}\mathbf{m})$	Irrigation $(\mathbf{m}\mathbf{m}/\mathbf{C}\mathbf{y}\mathbf{c}\mathbf{l}\mathbf{e})$	Cycle Duration (Days)	Plant Density $(\mathbf{P}\mathbf{l}\mathbf{a}\mathbf{n}\mathbf{t}\mathbf{s}/{\mathbf{m}}^2)$
Lettuce (Lactuca sativa)	15–20	300–500	6.0–7.0	1.4–1.8	280–560	45–60	10–20
Spinach (Spinacia oleracea)	15–25	300–500	6.0–7.0	1.6–1.8	400–500	50–70	20–25
Strawberry (Fragaria × ananassa)	15–25	400–600	5.3–6.5	1.0–2.0	400–500	90–100	10–20

3.3.3. Yield Estimation Approach

The theoretical yield estimation for each crop was derived from empirical data reported in national and regional technical manuals. The reference values considered were: 0.1–0.2 kg plant⁻¹ for lettuce, 0.1–0.25 kg plant⁻¹ for spinach, and 0.4–0.5 kg plant⁻¹ for strawberry [65,66,67]. These data served as input for the productivity model used in Section 4.2. to calculate economic performance indicators (NPV, IRR, and FCL).

To ensure physiological coherence, yield potential was expressed as a function of intercepted radiation, following the general formulation:

$$Y_p=RUE\times {PAR}_{abs}$$

(1)

where $Y_p$ is the potential yield $(\mathrm{g}/{\mathrm{m}}^2)$, RUE $(\mathrm{g}/\mathrm{M}\mathrm{J})$ is the radiation use efficiency specific to each crop, and ${PAR}_{abs}$ $(\mathrm{M}\mathrm{J}/{\mathrm{m}}^2)$ represents the absorbed photosynthetically active radiation (50% of total incident solar radiation). For partially shaded agrivoltaic conditions, a transmissivity factor (τ) of 0.72 was applied, following experimental data reported by Weselek et al. [10]. The RUE values were taken from controlled environment studies: RUE values adopted for modeling were 3.1 $\mathrm{g}/\mathrm{M}\mathrm{J}$ for lettuce, 3.8 $\mathrm{g}/\mathrm{M}\mathrm{J}$ for spinach, and 2.9 $\mathrm{g}/\mathrm{M}\mathrm{J}$ for strawberry, consistent with experimental reports for hydroponic and semi-shaded systems [63,64]. This approach ensured comparability among species under controlled light and temperature conditions and provided a reproducible framework to link crop productivity with the agrivoltaic system’s radiation regime.

3.4. Water Consumption Assessment

Water consumption analysis aimed to quantify the efficiency of the hydroponic subsystem integrated in the agrivoltaic (AV) configuration, comparing its performance against conventional irrigation practices commonly adopted in Colombian rural systems. The assessment followed a dual approach: (i) estimation of crop water requirements through reference evapotranspiration (ET_o) and crop coefficient (K_c) methodology, and (ii) computation of actual water demand under recirculating hydroponic operation, accounting for system losses by evaporation, percolation, and drainage.

Reference evapotranspiration (ET_o) values were derived from IDEAM meteorological stations in Pisba and Cabrera using the FAO-56 Penman–Monteith equation [36], based on daily temperature, relative humidity, wind speed, and global solar radiation. Annual mean ET_o ranged between 2.4 and 4.1 mm day⁻¹, depending on elevation and vegetation cover. Crop coefficients (K_c) were assigned for each phenological stage following national hydroponic manuals: lettuce (0.65–1.00), spinach (0.60–0.95), and strawberry (0.70–1.10) [65,66,67]. Total water requirements under open-field conditions were calculated as:

$${ET}_c={ET}_o\times K_c$$

(2)

Yielding seasonal consumptions between 420 and 550 mm cycle⁻¹ for traditional irrigation. These values align with agronomic baselines reported for temperate Andean zones [68,69]. The hydroponic subsystem employed a Nutrient Film Technique (NFT) configuration with recirculation and automated loss control, resulting in recapture efficiencies of 60–70% [70]. Daily nutrient solution volumes were determined from flowmeter readings and adjusted for evapotranspiration using:

$$W_{recirc}={ET}_c\times \left(1-E_r\right)$$

(3)

where E_r denotes the recovery efficiency. Under standard operation, the NFT modules required 1.2–1.5 L/plant/day, with total cycle consumption of 40–60 m³/ha, in contrast to 180–220 m³/ha estimated for open-field irrigation. This represents an overall reduction of 70–80%, consistent with reported AV-hydroponic savings in East Africa [9] and greenhouse-based WEFE frameworks [71].

The evaluation of water-use efficiency was conducted through the Water Use Efficiency (WUE) indicator, defined as the ratio between crop yield and the total volume of water applied during the cultivation cycle, expressed in kilograms per cubic meter (kg m⁻³). This metric allows comparison between production systems with different irrigation regimes or technologies. The theoretical expression used follows the formulation:

$$WUE={\displaystyle \frac{Y}{W}}$$

(4)

where Y is the total crop yield (kg/m), obtained from the productive model described in Section 3.3 and W is the cumulative water input (m³/m) during the cultivation cycle, including irrigation and nutrient-solution replenishment for the hydroponic configuration.

In the case of traditional irrigation, W corresponds to the cumulative evapotranspiration (ET_c) derived from the FAO-56 Penman–Monteith method, adjusted by the specific crop coefficient (K_c) and expressed as the water depth applied over the cultivated surface. For the hydroponic agrivoltaic system, W represents the effective volume of nutrient solution delivered and recirculated through the system, corrected by the recovery efficiency $\left(n_r\right)$ to account for evaporative and operational losses.

This formulation enables the quantification of the relative efficiency in water use between both production systems, establishing a dimensionless comparison that can later be correlated with energy consumption and photovoltaic generation in the integrated WEFSCC framework. By maintaining consistency with the FAO [47] guidelines and national technical manuals [65,66,67], this methodological approach ensures the reproducibility of calculations and provides a standardized basis for developing daily and annual water-consumption curves to be presented in subsequent sections. This formulation enables the quantification of the relative efficiency in water use between both production systems, establishing a dimensionless comparison that can later be correlated with energy consumption and photovoltaic generation in the integrated WEFSCC framework. By maintaining consistency with the Ref. [47] guidelines and national technical manuals [65,66,67], this methodological approach ensures the reproducibility of calculations and provides a standardized basis for developing daily and annual water-consumption curves to be presented in subsequent sections.

3.5. Energy Generation Estimation

The estimation of photovoltaic (PV) energy generation for the agrivoltaic (AV) system was conducted through a techno-environmental simulation framework that integrates (i) hourly solar resource reconstruction on the plane of array, (ii) PV module thermal-electrical performance, and (iii) system-level derating and availability. Historical time series of global solar radiation, sunshine duration, air temperature, and cloud cover were obtained from IDEAM stations for Pisba and Cabrera to ensure temporal consistency and gap filling. Sunshine duration was used to derive peak sun hours (PSH) and to characterize intra-annual seasonality typical of the Andean bimodal regime, which is necessary to generate daily and monthly production profiles.

3.5.1. Solar Resource Transposition to the Plane of Array (POA)

Hourly global horizontal irradiance (GHI) was decomposed into its beam (DNI) and diffuse (DHI) components using standard clearness-index correlations. The irradiance was then transposed from the horizontal to the tilted plane of the PV array to obtain the plane of array irradiance $G_{POA}$ as the sum of beam, sky-diffuse, and ground-reflected components:

$$G_{POA}\left(t\right)=B_{POA}\left(t\right)+D_{POA}\left(t\right)+R_{POA}\left(t\right)$$

(5)

where $B_{POA}=DNI·cos{\theta }_i$ (with ${\theta }_i$ as the incidence angle on the tilted plane), $D_{POA}$ is estimated with a sky-diffuse anisotropic model, and $R_{POA}={\rho }_g·GHI·{\displaystyle \frac{1-cos\beta }{2}}$ with ${\rho }_g$ as the ground albedo and $\beta$ as the tilt angle. For both sites, tilt and azimuth were set to maximize annual yield under fixed-tilt mounting; row spacing and ground coverage ratio (GCR) were selected to limit inter-row shading and to accommodate the hydroponic towers. The agrivoltaic structural shading was included through an hourly shading factor $f_{sh}\left(t\right)$ derived from the geometric layout (tower height, array elevation, row separation), so that $G_{POA}^{*}\left(t\right)=G_{POA}\left(t\right)·\left[1-f_{sh}\left(t\right)\right]$.

3.5.2. Cell Temperature and Module Power Model

Module cell temperature was computed from ambient conditions using an NOCT/NMOT-based relation:

$$T_{cell}\left(t\right)=T_{amb}\left(t\right)+{\displaystyle \frac{NMOT-20}{800}}{G}_{POA}^{*}\left(t\right)$$

(6)

with NMOT = 43 °C (as specified for the selected module class). The temperature coefficient of maximum power $\alpha P_{max}$ was applied to adjust power away from STC:

$$P_{dc}\left(t\right)=P_{STC}{\displaystyle \frac{G_{POA}^{*}\left(t\right)}{1000}}\left[1+\alpha P_{max}(T_{cell}\left(t\right)-25\right]$$

(7)

where $P_{STC}$ is the DC nameplate and $\alpha P_{max}$ is negative (manufacturer data). DC output was converted to AC using inverter efficiency curves and DC/AC sizing; hourly AC power is then:

$$P_{ac}\left(t\right)=P_{dC}\left(t\right){\eta }_{inv}\left(t\right)\prod \left(1-L_i\right)$$

(8)

3.5.3. Hourly to Monthly Energy Aggregation and Profiles

Hourly AC energy was computed as $E\left(t\right)=P_{ac}\left(t\right)∆t$ and aggregated to daily and monthly totals:

$$E_{day}=\sum _{h=1}^{24}E\left(t_h\right),E_{month}=\sum _{d=1}^D{E}_{day}\left(d\right)$$

(9)

where D denotes days in the month, and two families of curves are generated from the same simulation outputs (to be presented in Section 4): (i) daily profiles for typical days (e.g., the 15th of representative months), showing the pronounced midday peak in PV output; and (ii) monthly profiles across the year, evidencing seasonal variability driven by cloud cover, sun path, and tilt/orientation. These curves are time-aligned to the water-demand series from Section 3.4 (resampled to 30 min resolution) to analyze energy–water coupling at sub-daily and monthly scales).

Also, in Colombia, surplus electricity from distributed generation projects can be sold to the grid under the framework of Law 1715 of 2014 and its regulatory decrees, which promote the integration of renewable energy into the national matrix. The Energy and Gas Regulatory Commission (CREG) has established specific guidelines, Resolution 174 of 2021 [72] for grid interconnection and net billing (medición neta), enabling small-scale self-generators with renewable sources (≤1 MW) to inject surplus electricity into the distribution network. In rural electrification contexts, such as Pisba and Cabrera, this regulatory framework allows agrivoltaic projects to monetize surplus energy through credits applied to electricity bills or direct sales to local utilities (e.g., EBSA and ENEL-Codensa). These mechanisms not only ensure the economic viability of distributed solar projects but also strengthen the role of agrivoltaics in advancing Colombia’s rural energy transition and expanding access to clean energy.

3.6. Economic Analysis and WEFSCC Integration

The economic and systemic analysis of the agrivoltaic (AV) system integrated with vertical hydroponics was structured under a multidimensional evaluation framework based on the extended Water–Energy–Food–Soil–Climate–Communities (WEFSCC) nexus. This approach simultaneously incorporates technical, financial, social, and environmental indicators, allowing for an integrated assessment of resource interdependencies and resilience in vulnerable rural contexts [8,9]. The analysis goes beyond conventional return-on-investment methodologies by capturing synergies among productive efficiency, renewable energy generation, and social inclusion within post-conflict territories.

3.6.1. Boundary Conditions and Model Structure

The methodological design was implemented in a five-year dynamic cash flow model, developed in Microsoft Excel and parameterized with technical outputs from Section 3.4 and Section 3.5. It integrates hydroponic production performance, photovoltaic (PV) energy generation, and socio-economic indicators to simulate the systemic behavior of the AV unit (“Productive Unit”, PU). The boundary conditions defining the model are summarized below:

• Climatic parameters:
• Hourly series of solar irradiance, air temperature, and precipitation were obtained from IDEAM datasets, ensuring spatial representativeness for Pisba and Cabrera. Interannual variability of ±10–15% was used to represent climate uncertainty and to define upper and lower radiation limits for scenario evaluation.
• Agricultural parameters:
• Crop selection and yields were based on Section 3.3, considering phenological cycles, water-use efficiency (WUE), and the hydroponic recovery factor (ηr = 0.80–0.95). Nutrient solution losses, replacement frequency, and crop turnover rates were included to represent operational dynamics under variable radiation and temperature conditions [36].
• Energy parameters:
• PV generation capacity was fixed at 30 kWpDC, using a performance ratio (PR) of 0.79 derived from Table 1. The energy subsystem inputs included hourly irradiance (GPOA), temperature-adjusted module performance, and grid interconnection conditions following Law 1715 of 2014 and CREG Resolution 174 2021 [72]. Daily and annual generation curves were synchronized with water-demand cycles to analyze energy–water interactions within the WEFSCC model [35,53,55].
• Economic and financial parameters:
• The model incorporated investment (CapEx) and operation (OpEx) costs, including PV panels, inverters, structural components, hydroponic towers, and recirculation systems. Revenue streams included agricultural sales and surplus energy commercialization through net billing mechanisms. Key indicators are Net Present Value (NPV), Internal Rate of Return (IRR), and Payback Period, selected as primary financial outputs [13]. Sensitivity analyses were performed for discount rate, electricity tariff, equipment lifespan, and agricultural yields [28].

3.6.2. Scenario Configuration

Three simulation scenarios, conservative, moderate, and optimistic, were established to capture uncertainty in climatic, technical, and market conditions, ensuring a robust evaluation of system adaptability and resilience. These scenarios vary key parameters across environmental, technical, and financial dimensions, as summarized in Table 3.

Table 3. Definition of simulation scenarios and boundary conditions.

Parameter	Conservative	Moderate	Optimistic	WEFSCC Dimensions
Solar irradiance (kWh/m2/day)	−15% below long-term mean	Long-term mean	+10% above mean	Energy—Climate
Hydroponic recovery efficiency (ηr)	0.80	0.90	0.95	Water—Food
Crop yield (kg/m2/cycle)	−20%	Nominal	+15%	Food—Soil—Climate
Performance Ratio (PR)	0.75	0.79	0.83	Energy
Electricity tariff (USD/kWh)	0.07	0.09	0.11	Economy–Communities
Discount rate (%)	10	8	6	Economy
CapEx variation	+15%	Nominal	−10%	Economy
OpEx variation	+10%	Nominal	−5%	Economy–Communities
Community adoption factor 1	0.6	0.8	1.0	Communities–Climate

¹ The community adoption factor reflects the share of local stakeholders participating in operation and maintenance, influencing system sustainability and the social component of the WEFSCC nexus.

Basis for scenario ranges. The conservative–moderate–optimistic bands reflect observed and documented variability and uncertainty in the coupled agro-PV system and local markets. Solar irradiance was varied at −15%/mean/+10% to bracket interannual fluctuations in Andean sky conditions derived from the IDEAM time series used in the model (Section 3.5). The performance ratio (PR 0.75/0.79/0.83) follows the component-loss budget in Table 1 and published PR spreads for fixed-tilt systems of similar size under tropical conditions, accounting for temperature, soiling, mismatch, wiring, inverter efficiency, availability, and agri-shading. Hydroponic recovery efficiency (ηr 0.80/0.90/0.95) and crop yields (−20%/base/+15%) reflect (i) measured ranges for recirculating NFT systems and (ii) uncertainty in RUE-based yield under partial shade (transmissivity, temperature) for lettuce, spinach, and strawberry (Section 3.3). Financial levers (CapEx ±15/−10%, OpEx +10/−5%) capture supplier quotation dispersion and O&M intensity in rural settings; discount rates (10/8/6%) span typical hurdle rates for agri-energy community projects; tariffs (0.07/0.09/0.11 USD·kWh⁻¹) represent plausible rural net-billing bands under Colombia’s distributed generation framework referenced in Section 3.5. These ranges are thus traceable to the datasets, equipment specifications, and operating conditions documented in Section 3.3, Section 3.4 and Section 3.5, while ensuring robustness via sensitivity to the main physical and economic drivers.

3.6.3. Analytical Approach and Sensitivity Structure

Each scenario was executed under identical computational conditions, ensuring consistent comparison among results. Climatic, agronomic, and financial variables were varied parametrically to evaluate system performance and vulnerability. The conservative scenario represents unfavorable environmental and market conditions (reduced irradiance, lower yields, and higher costs), while the optimistic scenario simulates enhanced performance due to favorable climatic conditions and community engagement. The moderate scenario serves as the baseline configuration, corresponding to the mean historical conditions of both municipalities. The model applies a sensitivity analysis on critical variables (solar radiation, hydroponic efficiency, CapEx, OpEx, and discount rate) to quantify the marginal effect of parameter variation on financial and environmental indicators.

3.6.4. Integration Within the WEFSCC Framework

The three simulation scenarios enable the evaluation of multidimensional feedbacks among WEFSCC components: (i) Water–Energy: coupling between hydroponic recirculation cycles and PV generation profiles, ensuring system self-sufficiency. (ii) Energy–Climate: influence of radiation variability and temperature on PV output and overall performance ratio. (iii) Food–Soil: variation in yield efficiency and nutrient uptake under changing microclimatic conditions. (iv) Communities–Economy: local participation in system operation and energy monetization through cooperative schemes, improving territorial income distribution. (v) From the social dimension, each array of productive units (PU) is projected to generate approximately nine permanent jobs (production, operation, maintenance, commercialization), fostering inclusion of women and youth in local agricultural value chains [22,57]. (vi) From the environmental dimension, indicators include the reduction in water footprint, decrease in fossil fuel emissions, and land-use efficiency through dual productivity (energy and food) on the same surface, contributing to soil conservation and ecosystem restoration [3,73].

Finally, integrating these outcomes within the WEFSCC framework supports a systemic perspective of agrivoltaics as a resilient strategy for sustainable land management in post-conflict rural regions, linking climate adaptation, social reintegration, and economic viability [2,30,31].

3.7. Load Analysis

The structural assessment of the photovoltaic (PV) support tower was carried out to validate its mechanical stability under operational and extreme load conditions, ensuring compliance with Colombian and international design standards for light agro-industrial structures. The tower, hereafter referred to as the Productive Unit (PU), consists of modular ABS components assembled around four Guadua (Guadua angustifolia Kunth) columns that provide vertical support and structural stiffness. Each tower integrates five circular layers of ABS modules—each composed of four interlocking panels—stacked to reach the total height of the hydroponic system, while the PV panel is mounted at the top, supported by the same guadua framework (Figure 4).

Figure 4. Solar panel support tower assembly.

The load analysis followed a finite element analysis (FEA) approach using Autodesk Inventor^® 2024 with the FEA module, applying both static and dynamic simulations to determine the system’s stress distribution, displacement fields, and global safety factor (FS). The procedure consisted of defining the total acting forces on the tower (dead, live, wind, and operational loads) and distributing them among the five modules and four columns that constitute the base structure. The boundary conditions were defined to represent realistic support and environmental conditions: (i) Fixed constraints were applied at the column bases, replicating anchorage to the foundation plate. (ii) Gravity was included to account for the self-weight of structural components, hydroponic towers, and seedlings. (iii) External forces corresponded to wind pressure, hydroponic fluid loads, and the PV module weight. (iv) Contact and frictional constraints were implemented at the interfaces between ABS modules and guadua columns, ensuring appropriate load transfer.

For the ABS modules, mechanical properties were assigned directly from the software’s materials library (elastic modulus 2.3 GPa, yield strength 46 MPa, Poisson’s ratio 0.35). The Guadua angustifolia Kunth material was created as a custom orthotropic material following literature-reported properties (mean compressive strength 45 MPa, tensile strength 150 MPa, modulus of elasticity 12 GPa, density 730 kg m⁻³). Progressive mesh refinement was performed until the variation in the safety factor (FS) between iterations was below 20%, following recommended convergence criteria for structural validation in agrivoltaic applications [74]. The wind drag force acting on the tower and the solar module was calculated according to the standard aerodynamic formulation:

$$F={\displaystyle \frac{1}{2}}{C}_d·\rho ·A·v^2$$

(10)

where $C_d$ is the drag coefficient, $\rho$ the air density, A the projected area of the structure exposed to the wind (tower and solar panel with a 15° tilt), and v the wind velocity. For this case, the worst expected load condition was considered, corresponding to a wind velocity of 10 m/s at ground level, consistent with maximum registered speeds in open-field rural zones such as La Guajira. The air density was assumed at 1.25 kg/m³ (sea-level equivalent), and the drag coefficient $C_d=1.5$ was adopted for the combined tower–panel geometry, following ASCE-7 and CFE wind loading recommendations for cylindrical and planar hybrid structures. Table 4 summarizes the parameters used for the drag-force calculation.

Table 4. Parameter Values Used in the Drag Force Equation.

Parameter	Unit	Value
$C_d$	-	1.5
$\rho$	$\mathrm{k}\mathrm{g}/{\mathrm{m}}^3$	1.25
$A_{Tower}$	${\mathrm{m}}^2$	7.4
$A_{Panel}$	${\mathrm{m}}^2$	0.8
$F_{Tower}$	$\mathrm{N}$	773.2
$F_{Module}$	$\mathrm{N}$	128.2

The vertical loads include the self-weight of the structure (tower + PV module), hydroponic solution mass, and seedling biomass. The total static weight of the PV assembly was estimated at 706 N, distributed evenly across the four guadua columns. The hydroponic system load (water + plants) was estimated at 470.9 N, equivalent to 117.7 N per module, derived from the total mass of seedlings (48 kg). These values were used as service load conditions in the static FEA simulations (Table 5).

Table 5. Total Weight Supported by the Modules.

Parameter	Unit	Value
Seedlings Mass	kg	48
Seedlings Weight	N	470.9
Seedlings Weight/module	N	13.08

The load combination (wind + gravity + live load) was applied under both serviceability and ultimate limit states. The simulations followed the recommendations of the Colombian Seismic Resistant Code (NSR-10) and the AISC Design Guide 33 for hybrid composite structures. Mesh density was increased iteratively until convergence in stress and displacement results was achieved, verifying mesh independence. The safety factor (FS) was evaluated according to:

$$FS={\displaystyle \frac{{\sigma }_y}{{\sigma }_{max}}}$$

(11)

where ${\sigma }_y$ is the yield strength of the material and ${\sigma }_{max}$ is the maximum stress obtained in the FEA simulation. Convergence was confirmed when the change in FS between consecutive runs was below 20%. This procedure ensures structural reliability under both operational and extreme environmental conditions.

From the Soil–Climate perspective of the WEFSCC nexus, the load analysis contributes to the evaluation of environmental resilience by ensuring that the structural design minimizes land disturbance, avoids soil compaction, and withstands extreme climatic events. The use of Guadua angustifolia, a renewable native material, reinforces sustainability principles by reducing embodied carbon and promoting circularity in rural construction [8,9]. Additionally, the tower design enhances community adaptability and local fabrication potential, allowing replication using regionally available materials and labor, consistent with the Communities component of the WEFSCC framework. The validation of structural performance under realistic loading conditions thus guarantees that the AV system not only achieves technical feasibility but also supports ecological and social sustainability in post-conflict territories.

4. Results

4.1. Prototype Design

The structural validation of the agrivoltaic tower confirmed the mechanical stability and operational safety of the modular design under both service and ultimate limit conditions. The finite element analysis (FEA) demonstrated consistent convergence for the ABS modules and Guadua angustifolia Kunth columns, ensuring the reliability of the simulations and the adequacy of the mesh refinement process.

For the ABS base modules, the stress distribution results (Figure 5) revealed localized stress concentrations around the internal edges and perforations, while the majority of the structure remained within low-stress zones (<1 MPa). The maximum equivalent Von Mises stress reached 4.95 MPa, representing less than 11% of the material’s yield strength (46 MPa). The maximum displacement obtained was 0.94 mm, with an average safety factor (FS) of 4.0 (Table 6). These results confirm that the ABS geometry provides sufficient rigidity and deformation control to maintain dimensional stability under hydroponic and photovoltaic operational loads.

Figure 5. Finite element load simulation of the ABS base module. The color scale represents displacement (mm) under applied load. Grey circles indicate planting holes with no load/deformation, reserved for plant growth.

Table 6. Interactions for Different Validation Parameters.

Nodes	Von Mises (MPa)	Safety Factor	Displacement (mm)
20.275	0.43	15.0	0.46
101.748	1.16	15.0	0.53
107.015	0.72	15.0	0.53
134.401	0.98	15.0	0.53
310.109	3.87	5.2	0.94
433.687	3.57	5.6	0.94
774.689	4.37	4.6	0.94
794.010	3.95	5.1	0.94
1560.099	4.95	4.0	0.95

Similarly, the guadua support columns exhibited uniform stress distribution along their length, with peak Von Mises stresses concentrated near the fixed base and connection joints (Figure 6). The maximum stress value of 9.4 MPa corresponds to only 6.2% of the mean tensile strength of Guadua angustifolia (150 MPa), demonstrating a wide safety margin. The maximum displacement was approximately 10.05 mm, concentrated at the free upper end, consistent with slender structural behavior under combined axial and lateral loading. The minimum safety factor remained above 2.5 across all simulations (Table 7), fulfilling international and Colombian design standards for light hybrid structures (NSR-10, AISC-33).

Figure 6. Finite element simulation of Von Mises stress distribution in the guadua support column.

Table 7. Interactions for Different Validation Parameters Guadua Column.

Nodes	Von Mises (MPa)	Safety Factor	Displacement (mm)
13,725.00	9232	2.59	9.75
15,178.00	9327	2.57	10.04
69,327.00	9376	2.55	10.05
121,089.00	9375	2.56	10.05
517,616.00	9362	2.56	10.05
1,252,360.00	9405	2.55	10.05

The convergence of stresses, displacements, and safety factors in both components indicates that the model reached mesh independence, validating the numerical stability of the FEA process. The comparative analysis also showed that the load distribution between modules and columns was homogeneous, with minimal local instabilities or critical stress points. The modular configuration thus ensures efficient load transfer and adaptability to variations in hydroponic mass or photovoltaic weight, maintaining a balanced structural response.

From a design perspective, the prototype demonstrates optimal structural efficiency, combining lightweight ABS modules with renewable guadua columns to reduce embodied carbon and improve resilience to climatic variability. This result is consistent with recent agrivoltaic applications that integrate sustainable materials to reduce installation costs and environmental footprint [8,9,74]. The use of Guadua angustifolia provides additional benefits in flexibility, energy absorption, and vibration damping, increasing the system’s resistance to wind and seismic effects.

Overall, the structural results validate that the (PU) can withstand more than four times the design load without compromising safety or functionality. The integration of renewable and recyclable materials supports the environmental and community dimensions of the WEFSCC framework, linking structural integrity with local manufacturability and sustainable rural innovation. The validated design served as the physical and computational foundation for the subsequent evaluation of agricultural productivity, energy generation, and water-use efficiency.

4.2. Agricultural Productivity

The integration of hydroponic towers with photovoltaic panels within each Productive Unit (PU) produced a significant increase in land-use efficiency and overall agricultural yield across both municipalities. Each hydroponic tower consisted of five vertical levels with nutrient recirculation, designed to sustain controlled microclimatic conditions, reduce evapotranspiration, and maximize space efficiency. In total, 43 productive towers were simulated per site, each containing 20 plants per level (100 plants per tower), resulting in 4300 plants cultivated simultaneously per location. Three representative species were selected according to hydroponic feasibility, market relevance, and physiological adaptation to semi-shaded agrivoltaic conditions: lettuce (Lactuca sativa), spinach (Spinacia oleracea), and strawberry (Fragaria × ananassa).

Therefore, crop productivity was estimated using the radiation use efficiency (RUE) model defined in Equation (1) (Section 3.3.3), and ${PAR}_{abs}(\mathrm{M}\mathrm{J}/{\mathrm{m}}^2/\mathrm{d}\mathrm{a}\mathrm{y})$ represents the absorbed photosynthetically active radiation, calculated as 50% of the total incident solar radiation (GSR) multiplied by the system transmissivity $\left(\tau =0.72\right)$ to account for the partial shading produced by the photovoltaic modules [10]:

$${PAR}_{abs}=0.5\times GSR\times \tau$$

(12)

Based on historical data from the IDEAM, the average daily solar radiation was $4.1\mathrm{k}\mathrm{W}\mathrm{h}/{\mathrm{m}}^2/\mathrm{d}\mathrm{a}\mathrm{y}(14.76\mathrm{M}\mathrm{J}/{\mathrm{m}}^2/\mathrm{d}\mathrm{a}\mathrm{y})$ in Pisba and $4.3\mathrm{k}\mathrm{W}\mathrm{h}/{\mathrm{m}}^2/\mathrm{d}\mathrm{a}\mathrm{y}(15.48\mathrm{M}\mathrm{J}/{\mathrm{m}}^2/\mathrm{d}\mathrm{a}\mathrm{y})$ in Cabrera. Consequently, the absorbed PAR values were $5.31\mathrm{M}\mathrm{J}/{\mathrm{m}}^2/\mathrm{d}\mathrm{a}\mathrm{y}$ and $5.57\mathrm{M}\mathrm{J}/{\mathrm{m}}^2/\mathrm{d}\mathrm{a}\mathrm{y}$, respectively. The RUE coefficients, obtained from hydroponic studies under semi-shaded tropical conditions, were $3.1\mathrm{g}/\mathrm{M}\mathrm{J}$ for lettuce, $3.8\mathrm{g}/\mathrm{M}\mathrm{J}$ for spinach, and $2.9\mathrm{g}/\mathrm{M}\mathrm{J}$ for strawberry [63,64].

Applying Equation (1), the estimated daily potential yields $\left(Y_P\right)$ are shown in Table 8:

Table 8. Daily potential yields for Pisba and Cabrera.

Crop	RUE (g/MJ)	Yp Pisba (g/m2/Day)	Yp Cabrera (g/m2/Day)
Lettuce	3.1	16.47	17.28
Spinach	3.8	20.19	21.18
Strawberry	2.9	15.41	16.16

To convert these values into practical productivity metrics, the potential biomass per crop cycle was calculated using the typical growth periods for hydroponic cultivation (lettuce: 60 days, spinach: 50 days, strawberry: 90 days) (Table 9):

$$Y_{cycle}={\displaystyle \frac{Y_P\times duration}{1000}}$$

Table 9. Yield production for Pisba and Cabrera in kg.

Crop	Duration (Days)	Yp Pisba (kg/m2/Cycle)	Yp Cabrera (kg/m2/Cycle)
Lettuce	60	0.99	1.04
Spinach	50	1.01	1.06
Strawberry	90	1.39	1.45

With an average planting density of 10 plants·m⁻², the yield per plant ranged from 0.10 to 0.18 kg/plant per cycle for lettuce and spinach, and 0.4 for strawberry, consistent with the empirical data compiled in Colombian technical manuals [65,66,67]. The total yield per tower (PU) was then scaled to the 43 productive towers simulated per site, considering the number of annual harvests (lettuce: 7, spinach: 10, strawberry: 2) (Table 10).

Table 10. Total annual yield for Pisba and Cabrera.

Crop	Yield per Plant (kg) Pisba	Yield per Plant (kg) Cabrera	Harvests per Year	Annual Yield per PU (kg) Pisba	Annual Yield per PU (kg) Cabrera	Total Annual Yield Pisba (t)	Total Annual Yield Cabrera (t)
Lettuce	0.099	0.104	7	69.3	72.8	2.979	3.130
Spinach	0.101	0.106	10	101.0	106.0	4.343	4.558
Strawberry	0.139	0.145	2	27.8	29.0	1.195	1.247

The simulation incorporated three boundary scenarios consistent with the modeling framework established in Section 3.6, Conservative (−20%), Moderate (baseline), and Optimistic (+20%), reflecting variations in light availability, nutrient circulation efficiency, and temperature regimes, as shown in Table 11.

Table 11. Yield plus boundary scenarios for the production in Pisba and Cabrera.

Scenario	Lettuce (t/Year)	Spinach (t/Year)	Strawberry (t/Year)
Pisba—Conservative (−20%)	2.383	3.474	0.956
Pisba—Moderate (base)	2.979	4.343	1.195
Pisba—Optimistic (+20%)	3.574	5.211	1.434
Cabrera—Conservative (−20%)	2.504	3.646	0.997
Cabrera—Moderate (base)	3.130	4.558	1.247
Cabrera—Optimistic (+20%)	3.756	5.469	1.496

The results demonstrate that Cabrera outperformed Pisba by approximately 6–8% across all crops, primarily due to higher incident radiation and slightly reduced shading from the panel geometry. Spinach exhibited the highest radiation conversion efficiency (RUE = 3.8 g·MJ⁻¹) and the greatest total yield, while strawberry showed lower annual productivity because of its longer phenological cycle and reduced number of harvests.

From the perspective of the water–energy–food–soil–climate–communities (WEFSCC) nexus, the simulated results demonstrate that the agrivoltaic–hydroponic system enhances resource efficiency, climate resilience, and land productivity in mountain municipalities. By producing between 6 and 9 tons of hydroponic crops annually per hectare equivalent, the system increases the food output per unit area by over 100% compared to conventional open-field cultivation. The stable yield performance across both municipalities underlines the feasibility of deploying agrivoltaic technologies as a productive adaptation strategy for rural and post-conflict regions in Colombia.

4.3. Energy Generation

The estimation of photovoltaic (PV) energy generation for the agrivoltaic (AV) system in Pisba and Cabrera was conducted using the methodology described in Section 3.5, which integrates solar resource reconstruction on the plane of array (POA), PV module thermal-electrical performance, and system-level derating and availability. The hourly irradiance data were transposed to the tilted PV plane, corrected for shading due to hydroponic towers, and adjusted for temperature effects following the NOCT/NMOT model and manufacturer-provided temperature coefficients $\left(\alpha P_{max}\right)$ [9,13,42,43]. The DC power was subsequently converted to AC using inverter efficiency curves and the combined system losses described in Table 1 (product performance ratio PR = 0.79).

The resulting hourly AC power values were aggregated into daily and monthly totals to generate representative production profiles. Figure 7 presents the daily PV generation for a typical day of each representative month, illustrating the pronounced midday peak and the effect of partial shading. The PV generation curves for Pisba and Cabrera nearly coincide due to their identical installed capacity (30 kW) and similar irradiance conditions. The green line represents system consumption, which is almost constant throughout the day and significantly lower in magnitude—approximately 10% of the daily PV generation. Monthly aggregation (Figure 8) shows seasonal variation driven by solar incidence, cloud cover, and the Andean bimodal climate pattern. These curves were aligned with the water demand of the hydroponic system (Section 3.4) to evaluate the energy–water coupling at sub-daily and monthly scales.

Figure 7. Comparison between energy consumption and generation for a typical day of the month.

Figure 8. Comparison between energy consumption and generation for a year.

Table 12 summarizes key energy generation indicators for both sites. Pisba exhibited an average daily POA irradiance of 3.08 kWh/m²/day and a calculated cell temperature of 108.41 °C, resulting in 1.14 kWh per module per day of AC energy. Cabrera, with slightly higher irradiance (3.35 kWh/m²/day) and cell temperature (116.43 °C), achieved 1.18 kWh/module/day. Aggregated over 43 modules per site, the daily energy generation reached 49.07 kWh for Pisba and 50.95 kWh for Cabrera, corresponding to annual totals of 17.91 MWh and 18.60 MWh, respectively.

Table 12. Total annual energy generated for Pisba and Cabrera.

Municipality	Pisba	Cabrera	Units
Average Solar Radiation GSR	4.1	4.3	kWh/m2/day
AV Shading Factor fs	0.25	0.22
$\mathrm{POA}\mathrm{Irradiance}{G}_{POA}^{*}$	3.08	3.35	kWh/m2/day
$\mathrm{Cell}\mathrm{Temperature}{T}_{cell}$	108.41	116.43	°C
$\mathrm{DC}\mathrm{Power}\mathrm{per}\mathrm{Module}{P}_{dc}$	1.44	1.50	W/modulo/daily
$\mathrm{AC}\mathrm{Power}\mathrm{per}\mathrm{Module}{P}_{ac}$	1.14	1.18	W/modulo/daily
E daily, Array	49.07	50.95	kWh/day
E monthly, Array	1472.21	1528.47	kWh/month
E annual, Array	17.91	18.60	MWh/year

The total energy consumption of the AV-hydroponic system was evaluated considering the pumps, LED lighting, sensors, and central controller, as summarized in Table 13. The analysis indicates that the energy generated by the PV array is sufficient to meet the system’s daily demand (9.68 kWh/day), leaving a surplus available for local use or grid injection. This confirms the technical feasibility of the system under typical meteorological conditions and highlights the synergy between clean energy generation and sustainable agricultural production.

Table 13. Total energy consumption for the array.

Component	Power Rating (W)	Operation Time (h/Day)	Energy Consumption (kWh/Day)	Notes
Central irrigation pump	50	12	0.6	Serves all 43 towers
Water-level pump	50	24	1.2	Maintains solution level
Nutrient solution sensors	5	24	0.12	One set for the whole system
Central controller	5	4	0.02	One unit for the whole system
LED lighting	15	4	2.58	43 towers, 4 h/day each
Other actuators/valves	5	24	5.16	43 towers, includes pumps/valves
Total daily energy consumption			9.68	kWh/day
Total annual energy consumption			3.53	MWh/year

This section demonstrates the effective integration of PV energy generation with the hydroponic agrivoltaic system, confirming that the design provides sufficient power for operations while maintaining excess generation for potential community or grid applications. The results also provide the foundation for energy–water optimization in subsequent economic and sustainability analyses (Section 4.5).

4.4. Water Use Efficiency

The water consumption performance of the hydroponic subsystem integrated into the agrivoltaic (AV) Productive Units (PUs) was evaluated and compared with conventional open-field irrigation practices in Pisba and Cabrera. Each PU consists of a vertical hydroponic tower containing five levels, with 20 plants per level, totaling 100 plants per tower. Reference evapotranspiration (ET_o) values were obtained from IDEAM meteorological data and processed using the FAO-56 Penman–Monteith equation [36], with annual mean values ranging from 2.5 to 4.1 mm/day depending on elevation, vegetation cover, and local solar radiation. Crop coefficients (K_c) were applied according to phenological stage—lettuce (0.65–1.0), spinach (0.60–0.95), and strawberry (0.70–1.10)—resulting in crop evapotranspiration (ET_c) through ${ET}_c={ET}_o\ast K_c$.

Daily water requirements per plant were calculated by multiplying ET_c by the plant area, derived from the seeding density of 20 plants/m² (0.05 m² per plant). In the hydroponic AV system, water recirculation efficiency (n_r = 0.6–0.7) was applied to account for evaporation and drainage losses, yielding a daily water demand of 1.2–1.5 L/plant. Consequently, each PU requires 120–150 L/day, leading to an annual consumption of 43–55 m³, depending on crop type and seasonal variations. Conventional open-field irrigation, by contrast, demands 180–220 m³/year per PU, indicating a 70–80% reduction in water use under hydroponic AV conditions [68,69,70]. Water-use efficiency (WUE) was defined as the ratio of total crop yield to the cumulative water input over the cultivation cycle:

$$WUE={\displaystyle \raisebox{1ex}{$Y$}\!\left/ \!\raisebox{-1ex}{$W_{PU,annual}$}\right.}$$

where $Y$ is the annual yield per PU obtained from the model in Section 3.3 and $W_{PU,annual}$ is the total annual water consumed per PU. Table 14 summarizes the estimated WUE for each crop under the moderate scenario.

Table 14. Estimated Water Use Efficiency per Crop for a Single PU (100 plants).

Crop	Annual Yield per PU (kg) Pisba	Annual Water Use (m3)	WUE (kg/m3) Pisba	Annual Yield per PU (kg) Cabrera	Annual Water Use (m3/PU)	WUE (kg/m3) Cabrera
Lettuce	69.3	44	1.57	72.8	44	1.66
Spinach	101.0	43	2.35	106.0	43	2.47
Strawberry	27.8	55	0.51	29.0	55	0.53

The WUE values reflect the efficiency of the hydroponic AV system in maximizing yield while minimizing water input, particularly under partial shading generated by photovoltaic panels. Spinach exhibits the highest WUE due to its fast growth cycle and lower water requirements relative to biomass produced, explaining its selection as the primary crop for economic and productivity analysis.

Figure 9 presents the diurnal water consumption profile for a typical hydroponic cycle (6:00–18:00 h). Water uptake and recirculation rates followed the radiation-driven evapotranspiration pattern, showing midday peaks between 10:00 h and 13:00 h, when crop transpiration and nutrient absorption are highest. The synchronization of irrigation pulses with solar radiation improved the correlation between water and energy use efficiency within the WEFSCC framework.

Figure 9. Daily Hydroponic Water Consumption per PU (12 h cycle).

At the monthly scale (Figure 10), the AV-hydroponic configuration maintained stable water consumption despite climatic variability, contrasting with open-field irrigation systems that exhibit marked seasonality linked to precipitation and solar radiation. The cumulative hydroponic consumption ranged from 4–6 m³ month⁻¹ per crop, while conventional irrigation exceeded 18 m³ month⁻¹, aligning with the reductions observed in other agrivoltaic hydroponic studies in Mediterranean and equatorial climates [68,69,70].

Figure 10. Monthly Water Consumption Comparison.

Furthermore, the WUE analysis was conducted under three modeled scenarios (Section 3.6): Conservative (−20%), reflecting suboptimal nutrient circulation and limited light; Moderate (baseline), corresponding to average operational performance; and Optimistic (+20%), assuming enhanced nutrient solution management and extended photoperiods. The resulting WUE values demonstrate a clear response to management and climatic variability, providing a quantitative basis for integrating water efficiency metrics into the overall WEFSCC nexus analysis.

4.5. Financial, Economic, and Socio-Environmental Results

4.5.1. Municipality of Pisba

The financial, economic, and socio-environmental evaluation of the agrivoltaic (AV) pilot in Pisba was performed over a six-year projection period, expressed in constant 2025 U.S. dollars (USD 1 = 4004 COP) and discounted at an annual rate of 10%. The analysis followed international standards for rural infrastructure investments, allowing consistency with multilateral financing frameworks such as the Inter-American Development Bank (IDB) and the World Bank.

Investment Structure (CapEx): The total initial investment required for the deployment of the 43-tower hydroponic–photovoltaic array reached USD 30,759.74. As summarized in Table 15, 77% of this investment corresponds to the productive units (hydroponic towers and PV modules), 10% to installation and assembly, and another 10% to hydraulic and civil infrastructure (tanks, pumps, and foundations). Indirect costs associated with engineering design, permitting, and environmental licensing accounted for 5%, while an equivalent 5% contingency was included to cover potential overruns and exchange rate fluctuations. This configuration results in a unitary investment cost of approximately USD 715 per tower.

Table 15. Breakdown of Capital Expenditures (CapEx), Operating Expenditures (OpEx), and Revenues for the Agrivoltaic Pilot in Pisba.

Capital Expenditure (CapEx)					Operating Expenditure (OpEx)
Concept			Estimated (%)	Value (USD)	Concept		Estimated (%)	Value (USD)
Total Cost of (PUs)			-	$   23,661.34	Labor	Operators	-	$     8991
Installation and Assembly			10%	$    2366.13		Technicians		$     5994
Hydraulic/Civil Infrastructure			10%	$    2366.13	Preventive maintenance	Towers	0.5%	$    118.31
Indirect Costs (design, permits)			5%	$    1183.07		Sensors
Contingencies			5%	$    1183.07		Modules
Total Estimated CapEx			-	$   30,759.74		Hydraulics
Revenues (USD/year)					Gradual Replacement of Equipment	Sensors	0.5%	$    118.31
Source	Price per unit	Total production	Incomes			Pump
Electricity (kWh)	$     0.0652	17,911.89	$    1167.58	USD/Year		Modules
Lettuce (unit)	$     0.4163	16,328.22	$    6796.63	USD/Year	Agricultural inputs	Seeds	-	$    149.85
Strawberry (kg)	$     2.7473	1195.40	$    3284.07	USD/Year		Nutrients		$    149.85
Spinach (unit)	$     0.3996	36,191.67	$   14,462.20	USD/Year		Packaging		$    149.85
Total CapEx			$   15,629.79	USD/Year	Transport		-	$    299.70
					Non-Operating Expenses	Utilities & communications	-	$    149.85
						Local Transportation and Tickets		$    299.70
						Office supplies		$    449.55
						Safety gear		$    599.40
						Miscellaneous plant expenses		$    299.70
					Total OpEx	-	-	$    17,769

Operating Expenditures (OpEx): Annual operational costs were estimated at USD 17,769, largely driven by labor expenses (84%), comprising operators and specialized technicians responsible for hydroponic maintenance, fertigation calibration, and PV cleaning. Preventive maintenance of towers, sensors, and hydraulic systems accounted for 1–2% of OpEx, while agricultural inputs such as seeds, nutrients, and packaging totaled approximately USD 450/year. General administrative costs—utilities, transportation, safety equipment, and office supplies—amounted to USD 1800/year. This expenditure profile emphasizes the labor-intensive nature of the system, reinforcing its role in local job creation and technical capacity building, Table 15.

Revenue Structure: The annual revenue streams are diversified across agricultural and energy outputs. On the energy side, the 30 kWpDC PV array produced an estimated 17.9 MWh/year, generating USD 1167.6/year at a feed-in tariff of USD 0.0652/kWh under the national net-billing framework CREG Res. 174–2021 [72]. The hydroponic subsystem contributed the majority of income: spinach yielded USD 14,462.2/year, lettuce USD 6796.6/year, and strawberries USD 3284.1/year, totaling USD 15,629.8/year in agricultural revenues and energy generation, with 93% derived from agriculture and 7% from electricity.

Financial Performance: The free Cash Flow (FCL) analysis revealed a clear positive trajectory after the initial investment period (Figure 11). The project initiated with a capital outlay of −USD 33,324 in Year 0 and transitioned to positive cash flows from Year 1 (USD 499), increasing to USD 10,254 by Year 6. The internal rate of return (IRR) was calculated at 34.5%, with a discounted payback period of approximately five years and a cumulative net present value (NPV) of USD 11,500. This performance indicates robust profitability within the context of small-scale rural agrivoltaic systems. Even under conservative sensitivity conditions (−15% yield, +10% OpEx), the IRR remained above 22%, confirming the financial resilience of the pilot.

Figure 11. Pisba: FCL, Incomes and Social Results.

The parallel assessment of energy and spinach-only revenues evidenced that agricultural activities constitute the primary financial driver, while PV generation provides a steady baseline income contributing to liquidity stability. These results highlight the synergy within the WEFSCC framework, where energy production enhances agricultural resilience by powering the nutrient recirculation and monitoring systems.

Socio-Environmental Outcomes: Beyond its economic viability, the Pisba pilot yielded substantial environmental and social co-benefits (Figure 11). The initiative started with three direct jobs in Year 1 and progressively expanded to eight positions by Year 6, prioritizing local labor and women’s participation in agro-digital operations. The rainwater harvesting system achieved 1553 m³ in Year 1, accumulating over 6000 m³ across the 6-year horizon, exceeding hydroponic water demand by more than 200%. This performance avoided the extraction of approximately 3259 m³ of blue water, directly supporting Colombia’s National Water Policy targets [45]. In parallel, the renewable electricity generated by the PV array displaced ~763 t CO₂e during the evaluation period, calculated according to the emission factor for Colombia’s interconnected grid.

Sensitivity analysis: To ensure robustness, a scenario band analysis was applied in Pisba, incorporating ±20% variations in crop prices, ± 15% in agricultural yields, ±10% in electricity tariffs, and ±10% in both CapEx and OpEx. The generated band diagram (Figure 12) illustrates how the free cash flow evolves over the six-year horizon, with the central band (base case) showing a stable and progressive increase, while the lower band captures the potential effects of adverse shocks. Despite initial negative flows at year zero due to CapEx, the project recovers quickly and maintains positive cash flows across all bands from year one onward. The IRR never fell below ~22% in the worst-case configuration and reached values near 42% in favorable scenarios. This graphical evidence confirms that Pisba’s project maintains financial resilience under uncertainty, a critical condition for attracting green finance institutions.

Figure 12. Scenario band diagram of Free Cash Flow (FCL) for Pisba over a six-year horizon.

Risk and Resilience Considerations: The main project risks identified were crop price volatility, labor cost escalation, and maintenance uncertainties associated with remote Andean logistics. Mitigation strategies incorporated fixed-price supply contracts with institutional buyers, preventive maintenance scheduling, and a replacement fund equivalent to 9% of annual OpEx. This fund ensures short-term liquidity to address technical contingencies without compromising system continuity.

Overall, the Pisba agrivoltaic–hydroponic pilot demonstrates that integrated energy–food systems can achieve rapid capital recovery, stable cash flow generation, and measurable socio-environmental returns. These results validate the technical–financial model proposed and provide a reproducible benchmark for agrivoltaic scaling in high-altitude post-conflict regions of Colombia.

4.5.2. Municipality de Cabrera

The financial, economic, and socio-environmental evaluation of the agrivoltaic (AV) pilot in Cabrera was conducted over a six-year analysis period, expressed in constant 2025 U.S. dollars (1 USD = 4004 COP) and discounted at an annual rate of 10%. The assessment adopts a consistent methodological structure with the Pisba case, ensuring comparability across financial, technical, and environmental dimensions under the integrated WEFSCC framework (Water–Energy–Food–Soil–Climate–Communities).

Investment Structure (CapEx): The total capital expenditure for the Cabrera system reached USD 30,759.74, distributed similarly to the Pisba configuration. Productive units—comprising the 43 hydroponic towers, photovoltaic (PV) modules, and control systems—represented 77% of the total investment (USD 23,661.34). Installation and assembly accounted for 10% (USD 2366.13) (Table 16), while hydraulic and civil infrastructure (tanks, irrigation lines, and foundations) required an equivalent 10% (USD 2366.13). Indirect costs (engineering, design, permitting) and contingencies each represented 5% (USD 1183.07). This cost structure reflects efficient use of resources and a compact installation footprint. The resulting unit cost per productive unit (PU) was USD 715, identical to Pisba, underscoring standardization in system design and procurement. Compared with similar rural AV-hydroponic initiatives in the Andes, the CapEx remains 20–30% below reference levels.

Table 16. Breakdown of Capital Expenditures (CapEx), Operating Expenditures (OpEx), and Revenues for the Agrivoltaic Pilot in Cabrera.

Capital Expenditure (CapEx)					Operating Expenditure (OpEx)
Concept			Estimated (%)	Value (USD)	Concept		Estimated (%)	Value (USD)
Total Cost of (PUs)			-	$  23,661.34	Labor	Operators	-	$     8991
Installation and Assembly			10%	$   2366.13		Technicians		$     5994
Hydraulic/Civil Infrastructure			10%	$   2366.13	Preventive maintenance	Towers	0.5%	$    118.31
Indirect Costs (design, permits)			5%	$   1183.07		Sensors
Contingencies			5%	$   1183.07		Modules
Total Estimated CapEx			-	$  30,759.74		Hydraulics
Revenues (USD/year)					Gradual Replacement of Equipment	Sensors	0.5%	$    118.31
Source	Price per unit	Total production	Incomes			Pump
Electricity (kWh)	$     0.0652	18,596.38	$    1212.20	USD/Year		Modules
Lettuce (unit)	$     0.4163	17,152.88	$    7139.89	USD/Year	Agricultural inputs	Seeds	-	$    149.85
Strawberry (kg)	$     2.7473	1247.00	$    3425.82	USD/Year		Nutrients		$    149.85
Spinach (unit)	$     0.3996	37,983.33	$   15,178.16	USD/Year		Packaging		$    149.85
Total CapEX			$   16,390.36	USD/Year	Transport		-	$    299.70
					Non-Operating Expenses	Utilities & communications	-	$    149.85
						Local Transportation and Tickets		$    299.70
						Office supplies		$    449.55
						Safety gear		$    599.40
						Miscellaneous plant expenses		$    299.70
					Total OpEx	-	-	$    17,769

Operating Expenditures (OpEx): Annual operating costs for Cabrera were estimated at USD 17,769, maintaining a cost composition consistent with Pisba. Labor constitutes the main expense (84%), encompassing system operators, hydroponic technicians, and part-time maintenance personnel for the PV array. Preventive maintenance of towers, sensors, and hydraulic subsystems accounted for approximately USD 236/year (1%), and agricultural inputs (seeds, nutrients, packaging) reached USD 450/year. Non-operational costs—including transport, safety gear, communications, and administrative materials—amounted to USD 1800/year (Table 16). This operational profile reaffirms the project’s labor-intensive nature and its capacity to generate sustained rural employment. Moreover, the inclusion of women and young technicians in system operation supports the social dimension of the WEFSCC approach by fostering technical upskilling and community inclusion.

Revenue Structure: Cabrera’s revenue potential slightly exceeds that of Pisba due to higher incident solar radiation (4.3 kWh/m²·day) and increased agricultural productivity. The 30 kWpDC array produced approximately 18.6 MWh/year, generating USD 1212/year at the established feed-in tariff (USD 0.0652/kWh). Agricultural yields contributed the majority of income, led by spinach (USD 15,178/year), lettuce (USD 7139/year), and strawberry (USD 3426/year), resulting in total annual revenues of USD 16,390/year, 93% from crops and 7% from energy. These results validate the productivity model and illustrate the synergistic integration between the energy and agricultural subsystems, in which photovoltaic generation stabilizes production by powering irrigation pumps, nutrient dosing systems, and environmental sensors.

Financial Performance: The free cash flow (FCL) profile for Cabrera demonstrates a stronger financial outcome compared with Pisba (Figure 13). The project’s initial investment of −USD 32,919 (Year 0) was offset by rapid income growth, reaching USD 1060 in Year 1 and USD 11,320 by Year 6. The model produced a cumulative Net Present Value (NPV) of USD 18,700, an Internal Rate of Return (IRR) of 42%, and a discounted payback period below three years, confirming the pilot’s economic feasibility and scalability. The FCL analysis for the energy-only stream revealed constant negative flows (–USD 10,000/year on average), confirming that agricultural revenues are the main profitability driver. In contrast, the spinach-only scenario showed strong annual improvements, with FCL growing from USD 166 in Year 1 to USD 9620 by Year 6. This evolution underscores the central role of high-yield hydroponic crops under optimized water–energy conditions, as supported by comparable experiments in Kenya and southern Europe [9,27,33].

Figure 13. Cabrera: FCL, Incomes and Social Results.

Socio-Environmental Outcomes: From a socio-environmental standpoint, the Cabrera pilot replicates the positive externalities observed in Pisba, with higher magnitude due to superior climatic performance. The project began with three full-time jobs and expanded to eight by Year 6, ensuring gender inclusion and local capacity development. Rainwater harvesting achieved 1553 m³ in Year 1, increasing cumulatively to over 6000 m³ by Year 6, exceeding the water requirements of the hydroponic subsystem by more than sixfold. Renewable electricity generation mitigated approximately 789 t CO₂e over the six-year period, calculated using the Colombian grid emission factor. The integration of photovoltaic shading and hydroponic recirculation reduced open-field evapotranspiration losses by 70–80%, aligning with the results from agrivoltaic installations in Mediterranean climates [64]. The combination of water reuse, renewable energy, and high-yield cultivation demonstrates the closed-loop efficiency envisioned under the WEFSCC framework.

Sensitivity analysis. In Cabrera, the scenario band analysis revealed stronger resilience compared to Pisba. The band diagram (Figure 14) shows wider spreads between the worst and best scenarios, but with consistently higher cash flows across all years. Under the same uncertainty ranges (±20% in crop prices, ±15% in yields, ±10% in tariffs, CapEx, and OpEx), the base case maintains steady growth, while the optimistic band reaches nearly USD 70,000 by year six. The lower band reflects moderate risks but still remains positive after the initial investment period, avoiding critical financial stress. The IRR remained well above 25% under the worst-case scenario, while optimistic conditions pushed it beyond 45%. The diagrams highlight not only the robustness of the project but also its capacity to absorb variability in revenues and costs, making Cabrera particularly suitable for large-scale replication and commercial financing.

Figure 14. Scenario band diagram of Free Cash Flow (FCL) for Cabrera over a six-year horizon.

Risk and Resilience Considerations: Major risks identified include agricultural market fluctuations, energy tariff variations, and maintenance logistics in high-altitude terrain. To mitigate these, the project incorporated forward contracts for spinach and lettuce supply to local institutional programs, predictive maintenance scheduling, and a reserve fund equivalent to 9% of annual OpEx. Additionally, supplier agreements guarantee replacement of critical PV and hydraulic components within 72 h, ensuring operational continuity. The Cabrera pilot exhibits higher resilience to uncertainty compared with Pisba. Even under conservative conditions (−20% yield, +10% OpEx, +10% CapEx), the IRR remained above 25%, while optimistic scenarios projected IRR levels exceeding 45%. These findings confirm the project’s financial strength and its potential to attract climate finance instruments such as green bonds and blended finance schemes targeting rural decarbonization.

In summary, the Cabrera agrivoltaic system achieved superior technical and financial outcomes while replicating the social and environmental benefits observed in Pisba. With higher yields, enhanced solar resource availability, and robust profitability indicators, Cabrera stands out as a scalable demonstration site for agrivoltaic-hydroponic integration in Colombia’s highland municipalities. Its combination of rapid capital recovery, steady job creation, blue-water conservation, and emission mitigation illustrates how the WEFSCC nexus can translate into tangible socio-economic transformation in post-conflict rural areas.

5. Discussion

The comparative analysis of Pisba and Cabrera demonstrates that agrivoltaic (AV) systems integrating hydroponic cultivation and photovoltaic (PV) generation constitute a feasible model for sustainable land management in post-conflict Colombian regions. The results confirm that both sites achieve economic, environmental, and social viability, while revealing significant differences in system performance driven by microclimatic and socioeconomic conditions. These findings reinforce the conceptual foundations of the Water–Energy–Food–Soil–Climate–Communities (WEFSCC) nexus, highlighting the potential of decentralized hybrid infrastructures to address multidimensional rural vulnerabilities in the Andean context.

5.1. Water–Energy Coupling and Resource Efficiency

The synchronization between solar energy generation and water demand across a 12 h operational cycle (05:00–17:00) exhibited clear diurnal complementarity. Both Pisba and Cabrera presented peak energy outputs between 11:00 and 14:00, coinciding with the highest evapotranspiration and nutrient recirculation rates within the hydroponic towers. This alignment confirms the theoretical premises of resource synergy in AV systems described by Barron-Gafford et al. [13] and Geraldo et al. [31], where PV production stabilizes daytime energy availability for pumping and fertigation, reducing the need for energy storage. On an annual scale, the average energy generation reached approximately 17 MWh yr⁻¹ per site, consistent with modeled values for bifacial arrays in tropical latitudes reported by Riaz et al. [75] and CCST [17]. The distribution of monthly irradiance (derived from IDEAM actinometric datasets at 5.75° N, 73.0° W for Pisba and 4.5° N, 74.4° W for Cabrera) produced a typical bell-shaped curve with a ±12% amplitude, comparable to seasonal patterns in equatorial regions identified by Randle-Boggis et al. [9].

Hydroponic water demand was directly estimated from the crop evapotranspiration (ET_c = K_c × ET_o) following the FAO-56 methodology [36], integrated into the water balance of the nutrient film technique (NFT). The measured ET_c ranged between 1.7–3.2 mm day⁻¹ for lettuce and spinach and 2.8–4.1 mm day⁻¹ for strawberry. These values align with experimental data from agrivoltaic greenhouses reported by Yano and Cossu [53] and confirm that shaded microclimates under PV modules reduce water losses by 55–70% relative to open-field cultivation [13,61]. Consequently, the hydroponic towers achieved water use efficiencies (WUE) between 36 and 41 kg m⁻³, exceeding those observed in conventional irrigation by up to 300%. Such efficiency validates the argument of Mehta et al. [57] that coupling agrivoltaics with closed irrigation loops is a decisive adaptation strategy for developing countries facing water scarcity.

In Pisba, the harvested rainwater exceeded 5 times the annual hydroponic demand, while Cabrera achieved 4 times self-sufficiency. This confirms the robustness of the integrated design, consistent with circular resource management principles identified by Lucca et al. [3] and Gartsiyanova and Genchev [50]. The results thus substantiate the hypothesis that agrivoltaic–hydroponic integration optimizes both water productivity and climate resilience.

5.2. Agricultural Yields and Energy–Food Synergies

Crop performance exhibited site-dependent variations. The moderate elevation and higher mean temperature of Cabrera favored faster growth rates and slightly higher yields—104 g plant⁻¹ for lettuce, 106 g plant⁻¹ for spinach, and 145 g plant⁻¹ for strawberry—compared to Pisba’s 99–139 g plant⁻¹ range. The difference corresponds to a 4–6% increase in photosynthetic efficiency, consistent with the optimal light-partitioning dynamics described by Cossu et al. [55] and Adeh et al. [16], who noted that diffuse radiation under PV panels enhances canopy-level PAR utilization.

At the annual scale, Cabrera’s productive units (PUs) yielded ~29 kg of strawberries, 73 kg of lettuce, and 106 kg of spinach, while Pisba reached ~28 kg, 69 kg, and 101 kg, respectively. These results are coherent with comparable agrivoltaic trials on leafy vegetables in Mediterranean and tropical contexts [22,25,55]. The slightly superior productivity of Cabrera explains its higher operating margin (63% vs. 55%) and internal rate of return (IRR > 42%), while Pisba remained profitable with an IRR near 35%. The outcome supports the conclusion by Govinda et al. [26] that diversified AV portfolios combining electricity and horticultural revenues ensure financial sustainability even under conservative price scenarios.

5.3. Economic Performance and Financial Robustness

The financial analysis incorporated the complete cash flow over six years, distinguishing between agricultural and energy components. In both municipalities, capital expenditure (CapEx ≈ USD 30,760) was dominated by PV infrastructure (60%), hydroponic systems (30%), and civil works (10%). Operating costs (OpEx ≈ USD 17,770 yr⁻¹) were primarily associated with labor (82%) and preventive maintenance. Despite identical capital intensity, Cabrera’s cumulative free cash flow (FCL ≈ USD 11,320 yr⁻¹) was ~10% higher than Pisba’s (USD 10,254 yr⁻¹).

These margins are comparable to those achieved in integrated agrivoltaic farms in Europe and Asia, where unleveraged IRRs between 30 and 45% have been documented [22,23,33]. The main driver is the dual-income structure that stabilizes returns across agricultural and solar markets [8,26]. Moreover, both sites reached payback periods shorter than 3 years, confirming the model’s capacity to attract impact-oriented investment funds, as emphasized by Pascaris et al. [24] in the context of legal frameworks for agrivoltaic finance.

From a nexus perspective, the financial results also reveal the indirect benefits of water savings. By minimizing irrigation energy requirements and water procurement, both sites achieved cost reductions of ~18%, enhancing profitability without increasing land use. This mechanism exemplifies the economic interdependence described by Solano-Pereira et al. [48] and Chapagain et al. [2], who observed that feedback loops between water and energy inputs significantly amplify resilience in agricultural economies.

5.4. Environmental Impacts and Climate Dimension

The projects collectively avoided ~150 t CO₂e yr⁻¹, primarily through displacement of grid electricity by solar generation. The small difference between Pisba and Cabrera (2–3 t CO₂e) stems from marginal variations in irradiation and module temperature but remains negligible within the six-year total (≈ 0.9 kt CO₂e). This magnitude is consistent with life-cycle assessments of agrivoltaic installations worldwide [29,40]. Moreover, the AV shading reduced surface temperature by 2–4 °C and moderated soil moisture variability, mitigating thermal stress on plants—an effect also reported in drylands by Barron-Gafford et al. [13] and replicated in temperate regions by Weselek et al. [10]. Within the expanded WEFSCC lens, these results demonstrate tangible climate benefits that extend beyond carbon accounting, encompassing microclimatic regulation and adaptive capacity enhancement.

5.5. Social Inclusion and the “Communities” Nexus

Both pilot sites generated gradual employment growth from 3 to 8 positions within six years, involving local women and rural youth. The configuration of PUs allowed decentralized management by community cooperatives, reinforcing local governance structures in PDET/ZOMAC municipalities [38,39]. This outcome supports the findings of Khofi and Jewitt [27] and Mehta et al. [57], who identified community participation as a determinant of long-term sustainability in resource-nexus initiatives. In Pisba, the inclusion of female operators in crop management strengthened social cohesion, while Cabrera’s higher cash flow enabled training programs in solar maintenance and digital agriculture. These processes exemplify how agrivoltaic systems can act as peace-building instruments through employment, education, and empowerment, echoing the proposals of Chaher et al. [37] for circular and inclusive resource governance.

5.6. Comparative Synthesis and Policy Implications

Although Pisba and Cabrera differ in biophysical and socioeconomic characteristics, the unitary cost structure (Δ < 3%) validates the scalability and modularity of the system. Pisba’s configuration, optimized for high-rainfall zones, is better suited for publicly financed programs emphasizing water security and social impact. In contrast, Cabrera’s stronger profitability profile favors private investment schemes seeking rapid return on capital. This dual typology aligns with the hybrid financing model proposed by Geraldo et al. [31] and mirrors international strategies where agrivoltaic clusters are co-funded through green bonds and rural electrification funds. From a regulatory perspective, the evidence supports the introduction of fiscal incentives for AV technologies—such as accelerated depreciation of renewable assets, tax exemptions for decentralized microgeneration, and preferential credit lines for sustainable agriculture. These measures are consistent with policy frameworks in Europe [8,24] and could substantially accelerate adoption in Colombia’s post-conflict territories.

5.7. Methodological Limitations and Future Directions

The results presented derive from simulation-based modeling validated through technical coefficients and historical meteorological data rather than continuous field measurements. As noted by Giordano et al. [30] and Albrecht et al. [1], the robustness of nexus analyses depends on iterative feedback between models and empirical monitoring. Future stages will thus include a demonstration plot equipped with microclimate, radiation, and flow sensors to provide real-time validation of ET_c, energy output, and nutrient cycling.

Moreover, subsequent research should employ multi-criteria optimization approaches [58,76] to refine the balance among energy yield, crop productivity, and social return. The development of composite WEFSCC indicators—integrating climate vulnerability, soil health, and community resilience—would also enhance territorial planning capacities [3,49].

Finally, governance and financing innovations, such as community energy cooperatives and participatory ownership models, are recommended to ensure institutional durability and equitable benefit sharing, reinforcing the nexus between climate resilience and social reconstruction in rural Colombia.

6. Conclusions

The results of this study confirm the technical, economic, and socio-environmental feasibility of integrating agrivoltaic (AV) systems with hydroponic food production in Colombia’s post-conflict territories. By coupling renewable energy generation with intensive food cultivation under controlled water conditions, the proposed model demonstrates that dual land use can simultaneously enhance agricultural productivity, energy self-sufficiency, and local resilience. These findings reinforce the principles of the expanded Water–Energy–Food–Soil–Climate–Communities (WEFSCC) framework, which advocates for systemic optimization of resource flows across productive landscapes.

From a technical standpoint, the simulation of 30 kWpDC bifacial photovoltaic arrays in Pisba and Cabrera yielded an annual generation between 18 and 18 MWh, values consistent with comparable agrivoltaic pilots reported in Africa and southern Europe [8,9,13]. The thermal–electrical modeling and hourly transposition of solar radiation to the plane of array (POA) provided a robust estimate of generation peaks and seasonal variability, validated against historical IDEAM radiation datasets. This outcome confirms that the Andean region offers adequate solar potential for small-scale distributed generation, even under partial cloudiness typical of mountain climates. Moreover, the alignment between modeled energy profiles and empirical patterns reported in similar latitudes by Refs. [22,75] supports the internal validity of the simulation results.

On the agricultural side, the hydroponic subsystem exhibited significant water-use efficiency (WUE) advantages, reducing total consumption by 70–80% compared to conventional open-field irrigation, while maintaining yields of 2.9 t ha⁻¹ for lettuce, 4.5 t ha⁻¹ for spinach, and 1.2 t ha⁻¹ for strawberry per production cycle. These results coincide with global evidence on the synergistic performance of AV–hydroponic integrations [31,55]. The controlled recirculation of nutrient solution, adjusted to daily evapotranspiration derived from FAO-56 Penman–Monteith [36], demonstrated the capacity of the system to stabilize plant growth under reduced water input. Consequently, the AV configuration functions not only as a source of renewable energy but also as a microclimatic shield that mitigates water stress and increases photosynthetic efficiency, a key adaptation mechanism under ongoing climate change.

Financially, both municipalities displayed positive cash-flow trajectories and investment recoveries within five years, with unleveraged internal rates of return (IRRs) of 34.7% in Pisba and 42% in Cabrera. Sensitivity analysis across conservative, moderate, and optimistic scenarios revealed strong resilience to variations in crop price (±20%), yield (±15%), and tariff (±10%), confirming the robustness of the model to market and climatic uncertainty. These figures surpass standard profitability benchmarks for rural electrification projects and validate the hybrid AV–hydroponic model as a financially attractive pathway for decentralized renewable deployment in emerging economies.

From a social and environmental perspective, the pilots generated cumulative mitigation of ≈700 t CO₂e over six years, in addition to the creation of five permanent jobs per site, with progressive inclusion of women and rural youth. The projects also achieved water self-sufficiency through rainwater harvesting systems that exceeded irrigation requirements by more than 200%, thereby reducing pressure on blue-water resources. Such outcomes position agrivoltaics as a multifunctional strategy that simultaneously advances clean energy, climate adaptation, and local employment—objectives central to Colombia’s Planes de Desarrollo con Enfoque Territorial (PDET) and Zonas Más Afectadas por el Conflicto (ZOMAC) frameworks.

Despite the strong quantitative performance, several limitations must be acknowledged. The results are based primarily on techno-environmental simulations and financial modeling; therefore, long-term empirical validation under real operating conditions remains essential. Future work should include continuous monitoring of energy generation, water dynamics, and crop productivity using IoT-based sensors and data loggers to assess seasonal variability and long-term reliability. Additionally, the inclusion of a life-cycle assessment (LCA) and socio-economic impact evaluation would further substantiate the sustainability claims of the system.

From a policy and governance standpoint, the evidence supports the formulation of specific agrivoltaic regulatory frameworks within Colombia’s renewable-energy and rural-development agendas. Recommended measures include: (i) fiscal incentives for dual-use land systems; (ii) streamlined environmental licensing for low-impact AV installations; (iii) concessional financing mechanisms through national green funds and international climate-finance programs (e.g., GCF, IDB); and (iv) technical-capacity programs to train local operators in agro-digitalization, hydroponic fertigation, and PV maintenance. For international development agencies, priority should be given to co-financing modular pilots in post-conflict municipalities, where agrivoltaic systems can catalyze local economies while strengthening peacebuilding and community resilience.

In conclusion, the integration of agrivoltaics with hydroponic agriculture represents a scalable, climate-resilient, and socially inclusive solution for sustainable land management in Colombia’s Andean regions. The synergy observed between water efficiency, renewable energy generation, and agricultural yield demonstrates the operational viability of the WEFSCC nexus as a guiding framework for rural innovation. By merging technological precision with social participation, the model bridges productivity and peace, providing a replicable blueprint for achieving national goals of energy transition, food security, and territorial cohesion in post-conflict settings.

Furthermore, the extrapolation of the obtained results toward other Cundinamarca municipalities with comparable agroclimatic profiles such as Nemocón, Tocancipá and Susa, underscores the scalability potential of the proposed agrivoltaic–hydroponic con-figuration. These areas exhibit similar elevations (≈2500–2600 m.a.s.l.), temperature regimes (12–14 °C), and bimodal precipitation patterns to those of Pisba and Cabrera, suggesting strong replicability of both the technical and economic performance demonstrated in this study. Within the framework of the project “Microgrids: energía flexible y eficiente para Cundinamarca” (BPIN: 2021000100523), such territorial extrapolation aligns with the funding objectives of promoting decentralized renewable generation and sustainable food systems through pilot expansion in regions with analogous resource dynamics and agricultural frontiers.

This extension positions agrivoltaics as a modular and regionally adaptable strategy for sustainable rural development in Colombia’s Andean corridor, reinforcing the role of science-based planning in guiding policy and investment decisions under departmental innovation programs.