Agrivoltaic Systems for Scalable Dual-Use Solar: Configurations, Performance Trade-Offs, Economics, and Policy Pathways

1. Introduction

Rapid photovoltaic (PV) deployment is central to power-sector decarbonization because module performance, manufacturing scale, and system maturity have improved while costs have declined [1]. Utility-scale ground-mounted PV, however, converts land into a dedicated electricity-production asset. Where high-quality agricultural land, biodiversity, landscape values, and grid-accessible sites overlap, this conventional model can intensify land-use conflict and slow permitting.

Agrivoltaics, also described as agrophotovoltaics, agri-PV, or solar sharing, was proposed as a way to combine solar-energy conversion and crop cultivation rather than allocate land exclusively to one function [2]. Subsequent work established the technical potential of co-locating PV generation with crops, livestock, or other agricultural activity [3]. Recent reviews document a rapidly expanding body of experimental, modeling, environmental, and social research [4,5]. Field studies further show that partial shading can alter crop microclimate, soil temperature, and water use [6,7], while deployment in regions such as East Africa illustrates both climate-adaptation opportunities and infrastructure constraints [8].

The central analytical challenge is that AV performance cannot be represented by a single electricity metric. Module orientation, row spacing, mounting height, tracking strategy, crop canopy, irrigation, machinery access, and seasonal management jointly determine electricity yield and agricultural output. A design that maximizes annual kilowatt-hours per installed kilowatt may not maximize energy per hectare, crop value, land productivity, or grid value. Conversely, a high land equivalent ratio may conceal low electricity density, weak project finance, or inequitable distribution of benefits.

Existing reviews provide valuable agronomic and technological summaries, but three limitations remain important for solar-energy scholarship. First, configurations are commonly described case by case rather than compared under a common set of dimensions. Second, reported land-use and economic ranges are often repeated without their boundary conditions. Third, policy and equity are frequently discussed as general barriers instead of design variables that alter financing, agricultural continuity, and replicability. The present review addresses these limitations through a critical, boundary-conditioned synthesis aligned with the scope of PV systems, performance, deployment, and energy policy.

The review is organized around four research questions:

RQ1. How do elevated, spaced, tracking, vertical, and bifacial AV configurations differ in energy profile, agricultural compatibility, structural complexity, and evidence maturity?

RQ2. How should land-use efficiency and crop response be translated into variables and constraints that energy-system models can use?

RQ3. Under which technical, market, crop, financing, and policy conditions does AV become economically competitive?

RQ4. Which policy and ownership arrangements enable scalable deployment while protecting agricultural production and distributing benefits fairly?

The contribution is therefore not a claim that AV is universally preferable to conventional PV. It is a decision-oriented framework that distinguishes robust findings from context-dependent results, identifies where apparent contradictions arise from different baselines, and proposes a staged research program for representing AV within solar-resource assessment and energy-transition planning.

2. Review Methodology

2.1. Review Design and Scope

The revised article is presented as a critical integrative review. This design is appropriate because the evidence includes field experiments, crop trials, PV simulations, life-cycle assessments, techno-economic models, geospatial studies, legal analyses, and stakeholder research that cannot be combined through one common effect size. The review uses systematic elements an explicit search logic, eligibility rules, structured appraisal, and comparative extraction.

The unit of analysis is an AV study or authoritative policy/technical report that provides evidence relevant to at least one of five domains: (i) PV configuration and electricity performance; (ii) crop, microclimate, or land productivity; (iii) techno-economic performance; (iv) spatial or energy-system scalability; and (v) policy, ownership, social acceptance, or equity.

2.2. Search Channels, Search Logic, and Eligibility

Searches combined Google Scholar with publisher and indexing interfaces used to locate peer-reviewed records, including ScienceDirect, IEEE Xplore, SpringerLink, Wiley Online Library, and MDPI. Authoritative institutional material was sought from the International Energy Agency Photovoltaic Power Systems Programme, the European Commission Joint Research Centre, the U.S. Department of Energy, the National Energy Administration of China, and the Ministry of New and Renewable Energy of India. Backward and forward citation checking was used for foundational and frequently cited studies.

The core search expression was: ("agrivoltaic*" OR "agrophotovoltaic*" OR "agri-PV" OR "dual-use solar" OR "solar sharing") AND ("photovoltaic yield" OR bifacial OR tracking OR "land equivalent ratio" OR "crop yield" OR microclimate OR "techno-economic" OR LCOE OR policy OR equity OR ownership). Searches covered publications from 1982, when the co-location concept was formally proposed.

Eligible sources were written in English and directly addressed continued agricultural activity in combination with PV, or supplied an authoritative legal, policy, or energy-planning framework for such systems. Included evidence comprised peer-reviewed experiments, validated models, comparative reviews, geospatial analyses, and official reports. Exclusion criteria were conventional PV without an active agricultural output; commentary lacking a verifiable method or primary source; duplicate publications reporting the same result without additional analysis; and studies that did not report a configuration, outcome, or policy mechanism relevant to the review questions.

The earlier submission did not retain database-level hit counts. Accordingly, the revised article does not reconstruct an artificial PRISMA flow diagram or claim exhaustive database coverage. Instead, it reports transparent, reproducible search logic and treats the source cited corpus as an auditable critical-review evidence base. This limitation is stated explicitly again in Section 7.2.

3. Agrivoltaic Configurations and Energy Performance

AV design determines the distribution of irradiance between PV modules and the crop canopy. The same nominal capacity can produce different annual energy, hourly generation profiles, photosynthetically active radiation (PAR), machinery access, and structural loads depending on geometry and control. Crop-specific optimization studies therefore treat array design as a coupled food-energy problem rather than a conventional PV layout problem [9].

3.1. Elevated Fixed-Tilt and Tracking Systems

Elevated systems place modules several meters above the field, retaining clearance for machinery and enabling cultivation under the array. The German experience demonstrates that elevated layouts can preserve a large share of conventional PV yield while maintaining arable and horticultural operations, but the result depends on module density, orientation, and crop management [10,11]. Elevation can also moderate radiative loading and evapotranspiration, although the magnitude and agronomic value of this effect are climate- and crop-dependent [12].

Single-axis or dual-axis tracking adds a control variable: panel angle can be selected for electricity, crop light requirements, storm protection, or a combined objective. Comparisons between vertical and tracking bifacial systems show that annual yield alone does not capture the temporal and agricultural consequences of orientation [13]. Life-cycle assessment further indicates that additional steel, foundations, and structural height can offset part of the environmental benefit if material intensity is not controlled [14].

The principal engineering constraint is structural rather than photovoltaic. Greater height increases overturning moments, wind exposure, foundation requirements, and inspection burden. Tracking adds drives, sensors, communication, and failure modes. Consequently, evidence from a well-instrumented research installation cannot be transferred directly to a commercial farm without specifying span, height, design wind, soil bearing capacity, machinery envelope, and maintenance access.

3.2. Spaced, Row-Based, and Semi-Transparent Systems

Spaced systems use conventional or moderately elevated tilted modules with wider inter-row distances. Their advantage is constructability: existing mounting practice can be adapted, and open corridors can preserve farming access. Their principal penalty is lower installed capacity per hectare. French case evidence also shows that whether a project is perceived as a reconciliation of energy and agriculture depends on actual farming continuity, not only the presence of wider rows [15].

Spacing, tilt, and orientation jointly control direct-beam shadows and diffuse-light availability. Optimization of vertically and conventionally mounted arrays illustrates that reducing row distance can raise PV density while sharply reducing crop light, so the optimum depends on whether the objective is energy per installed kilowatt, electricity per hectare, crop yield, or combined value [16]. Bifacial row-based concepts may recover part of the density penalty through rear-side irradiance, but the gain depends on ground reflectance and obstruction [17].

Semi-transparent and greenhouse-integrated systems distribute light through module spacing or partial optical transmission. They are promising for high-value horticulture and controlled environments, but they should not be generalized to open-field arable systems because the crop value, support structure, ventilation, and reference greenhouse fundamentally change the economic boundary.

3.3. Vertical Bifacial Configurations

Vertical bifacial arrays are commonly oriented east-west, exposing one face to morning irradiance and the other to afternoon irradiance. Relative to optimally tilted monofacial PV, this geometry often reduces the midday peak and can create two broader generation peaks. The profile may be valuable where morning and evening electricity have higher system value or where midday curtailment is material. Module-level comparisons show that vertical bifacial performance must be assessed with both front- and rear-side irradiance rather than with nameplate capacity alone [18].

The effective bifacial gain is governed by module bifaciality, row-to-height ratio, horizon and mutual shading, crop-canopy height, seasonal albedo, snow cover, and mismatch across cells and strings. Crop and soil surfaces are dynamic reflectors: bare soil, dense foliage, senescent vegetation, and snow can produce materially different rear irradiance. Computational-fluid-dynamics and microclimate modeling also indicate that vertical rows modify wind and temperature fields differently from overhead structures [19].

Vertical rows occupy little physical footprint at the pile line and facilitate machinery movement parallel to the array. Nevertheless, land occupation cannot be represented only by foundation area because operational exclusion zones and row spacing determine usable agricultural area. High-latitude evidence shows that vertical bifacial systems can be technically attractive when rows are sufficiently separated and seasonal albedo supports rear-side generation; one recent study found feasible crop-PV performance at row separations above approximately 8 m and showed that crop albedo can materially affect electricity yield [20]. These findings are configuration-specific rather than a universal vertical-PV advantage.

The strongest case for vertical bifacial AV is therefore not necessarily maximum annual kilowatt-hours per module. It is the combined value of a distinct hourly profile, limited overhead obstruction, reduced soiling or snow retention in some climates, accessible farming corridors, and potential rear-side gains. Conversely, low albedo, narrow rows, high latitude-season mismatch, or poor electrical matching can erase the expected benefit.

3.4. Cross-Configuration Comparison

Table 1. Unified comparison of principal agrivoltaic configurations. Ratings are directional and must be interpreted with the boundary conditions in the final column.

Configuration	PV output/profile	Agricultural compatibility	Structural/operational burden	Typical economic implication	Critical boundary conditions
Elevated fixed-tilt	High annual yield potential; conventional daytime profile.	High clearance; suitable for machinery and diverse crops.	High foundations, steel, wind loading, and inspection burden.	Higher CAPEX may be offset by high-value crops and preserved land use.	Height, wind zone, soil, ground coverage ratio, crop and machinery envelope [10,11,14].
Elevated tracking	Potentially high yield and controllable shade; actuator losses and downtime possible.	Dynamic light management, but moving structures constrain operations and safety.	Highest mechanical and control complexity.	Value depends on energy-price profile and ability to monetize crop-light control.	Tracking algorithm, stow strategy, maintenance, crop phenology [13,21].
Spaced tilted rows	Lower capacity per hectare; familiar PV performance per module.	Open corridors and incremental adaptation of conventional farms.	Moderate structure; relatively simple maintenance.	Lower structural premium but opportunity cost from reduced PV density.	Row spacing, tilt, field shape, machinery direction, shadow length [15,16].
Vertical bifacial	Morning/evening peaks; rear-side contribution; often lower noon output.	Minimal overhead obstruction and parallel machinery access.	Moderate structure; requires bifacial modeling and careful electrical design.	Can gain value from profile, albedo, and low land obstruction.	Latitude, row-to-height ratio, albedo, snow, canopy, bifaciality, mismatch [18,19,20].
Semi-transparent/greenhouse	Transmission-yield trade-off; environment is partly controlled.	Best suited to protected horticulture and high-value crops.	Integrated building/greenhouse structure and climate control.	Crop value and avoided greenhouse costs may dominate electricity economics.	Optical transmission, crop light saturation, ventilation, greenhouse reference [5,22].

Table 1. Unified comparison of principal agrivoltaic configurations. Ratings are directional and must be interpreted with the boundary conditions in the final column.

Configuration	PV output/profile	Agricultural compatibility	Structural/operational burden	Typical economic implication	Critical boundary conditions
Elevated fixed-tilt	High annual yield potential; conventional daytime profile.	High clearance; suitable for machinery and diverse crops.	High foundations, steel, wind loading, and inspection burden.	Higher CAPEX may be offset by high-value crops and preserved land use.	Height, wind zone, soil, ground coverage ratio, crop and machinery envelope [10,11,14].
Elevated tracking	Potentially high yield and controllable shade; actuator losses and downtime possible.	Dynamic light management, but moving structures constrain operations and safety.	Highest mechanical and control complexity.	Value depends on energy-price profile and ability to monetize crop-light control.	Tracking algorithm, stow strategy, maintenance, crop phenology [13,21].
Spaced tilted rows	Lower capacity per hectare; familiar PV performance per module.	Open corridors and incremental adaptation of conventional farms.	Moderate structure; relatively simple maintenance.	Lower structural premium but opportunity cost from reduced PV density.	Row spacing, tilt, field shape, machinery direction, shadow length [15,16].
Vertical bifacial	Morning/evening peaks; rear-side contribution; often lower noon output.	Minimal overhead obstruction and parallel machinery access.	Moderate structure; requires bifacial modeling and careful electrical design.	Can gain value from profile, albedo, and low land obstruction.	Latitude, row-to-height ratio, albedo, snow, canopy, bifaciality, mismatch [18,19,20].
Semi-transparent/greenhouse	Transmission-yield trade-off; environment is partly controlled.	Best suited to protected horticulture and high-value crops.	Integrated building/greenhouse structure and climate control.	Crop value and avoided greenhouse costs may dominate electricity economics.	Optical transmission, crop light saturation, ventilation, greenhouse reference [5,22].

3.5. Configuration-Specific Research Needs

The priority is not another generic call for long-term research, but a configuration-specific test protocol. Elevated systems require reporting of height, span, foundation mass, wind design, and machinery clearance. Spaced systems require row spacing, ground coverage ratio, shadow distribution, and lost capacity per hectare. Vertical bifacial systems require measured front/rear irradiance, seasonal albedo, bifaciality, row-to-height ratio, string layout, and mismatch losses. Tracking systems require control logic, stow events, actuator energy, downtime, and maintenance.

Three targeted gaps are especially important for solar engineering. First, rear-side yield models should represent crop-canopy development rather than assume constant ground albedo. Second, hourly grid value should be evaluated alongside annual yield, particularly for vertical and tracking systems. Third, standardized paired reporting kilowatt-hours per kilowatt, kilowatt-hours per hectare, PAR at crop level, crop yield and quality, and structural material per kilowatt would allow design comparisons without collapsing distinct objectives into one indicator.

4. Land-Use Efficiency and Agricultural-Energy Interactions

The central land-use claim of AV is that one hectare can provide two outputs that would otherwise require separate areas. This claim is plausible but highly sensitive to the definition of the reference systems. The original land equivalent ratio (LER) formulation established a useful combined-productivity measure [23], while subsequent UK and northern-European assessments show that technical potential depends on crop choice, geometry, and the conventional PV comparator [24,25,26].

4.1. Land Equivalent Ratio: Definition, Strengths, and Limits

The land equivalent ratio is calculated by adding two normalized components: crop yield under AV relative to a matched crop-only reference and electricity generation under AV relative to a matched PV-only reference.

A value above one indicates that the combined crop-PV system produces more normalized output on an equivalent land area than the two segregated reference systems. Because this interpretation depends on the selected crop-only and PV-only baselines, both component ratios and the reference land areas should be reported explicitly [23,27,28].

Selected studies report LER values roughly between 1.1 and 1.7, with higher values generally occurring where moderate shading preserves crop yield while PV output remains substantial [6,10,25,29]. This is not a pooled effect estimate. The range combines different crops, climates, PV reference densities, temporal periods, and accounting rules. A value calculated against a dense conventional solar farm is not directly comparable with a value calculated against a low-density reference array. Likewise, support-footprint accounting can produce a different result from whole-project-area accounting.

LER also lacks three attributes needed by energy planners. It does not represent the timing or grid value of electricity, does not value crop quality or revenue, and does not reveal whether agricultural production remains primary or merely nominal. Therefore, LER should be reported with its two component ratios, the reference areas, annual energy density, crop-quality indicators, and uncertainty. Co-benefit frameworks can then extend the assessment to water, ecosystem, and social services without embedding them invisibly in one score [30].

4.2. Microclimate and Crop-Specific Responses

PV modules alter short-wave radiation, long-wave exchange, wind, humidity, soil temperature, and evapotranspiration. Field evidence shows that partial shade can conserve soil moisture and improve water-use efficiency under hot, dry conditions [29,31]. Microclimate models help explain spatial gradients beneath and between rows [32]. In tropical field trials, shading improved microclimate and performance for selected mung-bean genotypes, but the result should not be generalized beyond similar crop and climate conditions [33].

Crop response is governed by light saturation, phenology, water stress, temperature, and disease pressure. Reviews identify leafy vegetables, some forage crops, and selected legumes as comparatively shade-tolerant [34,35]. Controlled shading experiments show that potatoes can exhibit quality and yield responses that vary with shading level [36], while alfalfa studies demonstrate delayed flowering and reduced reproductive growth under shade [37]. Apple research similarly shows that shading may change fruit quality, not only total yield [38]. Partial-transparency studies with strawberries further demonstrate that the spatial uniformity of light matters in addition to its total quantity [22].

These findings explain why apparently conflicting results are not necessarily contradictory. In a water-limited system, reduced radiation can relieve heat and moisture stress; in a cool or light-limited system, the same reduction can constrain photosynthesis and delay development. AV performance should therefore be evaluated against the dominant local constraint. Crop selection and dynamic management are legitimate design variables, but post hoc crop substitution should not be used to justify a system that displaces the farm activity for which the land was approved [39,40].

4.3. Translating Land Efficiency into Energy-System Planning

Energy-system models require physical capacity and generation inputs rather than LER alone. Spatial assessments can estimate AV potential by combining agriculturally and legally eligible land, a realistic deployable share, configuration-specific capacity density, and the corresponding location-specific capacity factor.

The eligible land estimate should be reduced to account for ownership, social acceptance, crop suitability, biodiversity protection, legal restrictions, and grid-access constraints. LER should then enter the model as an agricultural-performance constraint or co-objective, such as a minimum crop-yield-retention requirement, rather than as a direct multiplier of electrical capacity [41].

This distinction has quantitative consequences. A Joint Research Centre assessment estimated that deploying AV on 1% of EU utilised agricultural area could accommodate approximately 944 GWdc at an assumed AV density near 0.6 MW about half the capacity density of conventional ground-mounted PV on the same land, but still above the EU 2030 solar target used in that assessment [41]. The result is a scenario illustration, not a forecast: eligible land, adoption, grid connection, crop safeguards, and public acceptance would reduce the technically calculated potential. Nevertheless, it shows how land eligibility and capacity density can alter energy-transition pathways even when AV produces fewer watts per hectare than a dedicated solar farm.

Capacity-expansion models should therefore represent several AV archetypes rather than one generic technology. Each archetype needs capital cost, fixed and variable operation cost, hourly generation profile, capacity density, eligible-land class, crop-yield constraint, water or adaptation co-benefit where defensible, and interconnection cost. Vertical bifacial and tracking archetypes may have system value that annual energy metrics miss, while elevated systems may have lower capacity density but broader crop compatibility.

4.4. Representative Performance Evidence

Table 2. Representative evidence and the boundaries required to interpret land and crop performance.

Evidence context	Configuration/crop	Reported contribution	Comparator/boundary	Appraisal and transferability
France, foundational field design [23]	Elevated PV with crops.	Established LER-based combined land productivity.	Separate crop and PV references.	Strong conceptual comparator; site and crop specific.
Germany [10]	Elevated arable/vegetable AV.	Demonstrated joint energy and agricultural operation.	German climate, pilot geometry, matched references.	Moderate-to-strong; detailed but not universally transferable.
Dryland USA [29]	PV shade with food-energy-water monitoring.	Mutual microclimate, water, crop, and PV benefits.	Hot, water-limited environment.	Strong field evidence for similar dryland constraints.
Northern latitudes/Sweden [26,42]	Vertical and other northern AV designs.	Validated performance and economic modeling under high-latitude conditions.	Seasonal irradiance and local crop systems.	Moderate; valuable for northern climates, scale remains limited.
UK synthesis [24,25]	Crop-based configurations and demonstration sites.	Shows technical/economic potential and variability among comparable sites.	UK climate, crop and market assumptions.	Moderate; emphasizes need for consistent configuration data.
Nigeria [33]	Shaded mung-bean genotypes.	Improved microclimate and selected crop responses.	Tropical genotype-specific field trial.	Context-specific; supports mechanism, not a universal crop claim.
Potato, alfalfa, apple, strawberry [22,36,37,38]	Controlled shade or AV-related crop studies.	Yield, phenology, quality, and light-uniformity effects vary by crop and stage.	Crop-specific experimental baselines.	Strong for mechanisms; limited direct transfer across crops.

Table 2. Representative evidence and the boundaries required to interpret land and crop performance.

Evidence context	Configuration/crop	Reported contribution	Comparator/boundary	Appraisal and transferability
France, foundational field design [23]	Elevated PV with crops.	Established LER-based combined land productivity.	Separate crop and PV references.	Strong conceptual comparator; site and crop specific.
Germany [10]	Elevated arable/vegetable AV.	Demonstrated joint energy and agricultural operation.	German climate, pilot geometry, matched references.	Moderate-to-strong; detailed but not universally transferable.
Dryland USA [29]	PV shade with food-energy-water monitoring.	Mutual microclimate, water, crop, and PV benefits.	Hot, water-limited environment.	Strong field evidence for similar dryland constraints.
Northern latitudes/Sweden [26,42]	Vertical and other northern AV designs.	Validated performance and economic modeling under high-latitude conditions.	Seasonal irradiance and local crop systems.	Moderate; valuable for northern climates, scale remains limited.
UK synthesis [24,25]	Crop-based configurations and demonstration sites.	Shows technical/economic potential and variability among comparable sites.	UK climate, crop and market assumptions.	Moderate; emphasizes need for consistent configuration data.
Nigeria [33]	Shaded mung-bean genotypes.	Improved microclimate and selected crop responses.	Tropical genotype-specific field trial.	Context-specific; supports mechanism, not a universal crop claim.
Potato, alfalfa, apple, strawberry [22,36,37,38]	Controlled shade or AV-related crop studies.	Yield, phenology, quality, and light-uniformity effects vary by crop and stage.	Crop-specific experimental baselines.	Strong for mechanisms; limited direct transfer across crops.

4.5. Decision Implications

A defensible land-use decision should report at least five outputs together: annual electricity per installed kilowatt, annual electricity per project hectare, crop yield and quality relative to a matched reference, the two components of LER, and the distribution of PAR or shade across the field. This prevents a high electricity yield per module from concealing low land productivity and prevents a high LER from concealing a weak or poorly timed electricity contribution.

5. Techno-Economic Performance and Scalability

AV economics are conditional because the project produces multiple outputs and introduces configuration-specific costs. The revised synthesis therefore avoids treating a 5–30% cost premium or a 7–12-year payback period as universal. Such ranges combine different heights, tracking systems, installed capacities, crops, electricity tariffs, financing structures, and subsidies. The relevant question is not “What is the AV payback period?” but “Under which explicitly defined boundary conditions is a particular AV design competitive?”

5.1. Cost Architecture by Configuration

Elevated AV commonly increases structural steel, concrete, geotechnical work, erection time, design certification, and electrical routing. The German techno-economic analysis demonstrates that price-performance is inseparable from policy and system design [43]. Indian scale-up analysis similarly identifies finance, standardization, land rules, and institutional coordination as major cost determinants beyond the PV hardware itself [44].

Spaced systems may use familiar structures but sacrifice installed capacity per hectare and can increase cabling or access-road length per megawatt. Vertical bifacial systems reduce overhead structure but require bifacial modules, accurate rear-side modeling, foundations designed for exposed rows, and layouts that maintain agricultural corridors. CFD and northern-field evidence show that geometry also affects local wind and microclimate, which can influence both structural design and crop performance [19,42].

Soft costs can be as important as hardware: agricultural monitoring, specialized permitting, interconnection studies, land and crop contracts, liability allocation, insurance, and coordination between a solar operator and a farm business. These costs are likely to decline through standardization, but only if standards preserve meaningful agricultural activity rather than reclassify conventional solar farms through nominal planting.

5.2. Multi-Output Valuation

LCOE remains useful for comparing the electricity component, but it cannot by itself evaluate a dual-output system. A complete investment model should report electricity LCOE together with project NPV, internal rate of return, discounted payback, agricultural gross margin, landowner and farmer cash flows, and sensitivity to yield and price [27,28].

Total project value should combine discounted electricity and agricultural cash flows, deduct AV-specific incremental costs, and include only co-benefits that are monetized and supported by evidence. Water savings, resilience, and ecosystem services should otherwise be reported separately. Economic-environmental analyses show that joint valuation can change the ranking of AV and segregated land uses [27], while apple-farming analysis demonstrates that crop-specific synergies can reduce the effective cost of AV under suitable conditions [28].

Agricultural income can diversify project revenue, while electricity income can buffer crop volatility. Weather-risk research supports the value of diversification to landowners [45]. However, this benefit is not automatic: electricity and agricultural risks may be correlated through heat, drought, storm, or equipment damage, and contract structures may allocate upside and downside asymmetrically between the developer, landowner, tenant farmer, and labor force.

5.3. Boundary-Conditioned Economic Evidence

Table 3. Techno-economic evidence interpreted through project boundaries rather than universal ranges.

Study/context	Configuration and scale	Metric or finding	Market/policy boundary	Agricultural boundary	Transferable interpretation
Germany [43]	Elevated pilot and modeled scale-up.	Higher structure cost evaluated against combined performance.	German tariffs, policy, engineering and financing assumptions.	Continued arable production.	Policy and design can offset cost, but result is not a generic CAPEX premium.
Niger [46]	Modeled food-energy case.	Economic feasibility assessed in a land- and resource-constrained setting.	Local energy access, prices and financing.	Crop income and food-energy context.	AV value can be high where energy access and land co-use dominate conventional LCOE comparisons.
Europe [47,48]	Geospatial/configuration comparisons.	Capacity density and design strongly alter technical-economic potential.	European land, irradiance and cost assumptions.	Arable-land suitability and access.	Regional potentials require archetype-specific density and eligibility.
Sweden/northern systems [26,42]	Experimental/model validation.	Economics tied to vertical geometry, northern irradiance and local agriculture.	Swedish prices and seasonal generation.	Northern crop calendar and machinery.	High-latitude evidence is valuable but should not be transferred to southern markets unchanged.
Sweden, 47 ha vertical case [49]	Farmer- and investor-led business models.	Reported farmer-led NPV ≈ EUR 5.5 million, ROI 8.3%, discounted payback 15.1 years.	Case-specific electricity price, discount rate, scale and contracts.	Crop profitability and value allocation differ by ownership model.	Profitability and equity can diverge; the business model is part of system design.
Systematic evidence [50]	Multiple crops, countries and configurations.	Productivity and profitability vary widely across assumptions.	Heterogeneous incentives and markets.	Different yield baselines and crop values.	Report distributions and sensitivities; do not present one payback benchmark.

Table 3. Techno-economic evidence interpreted through project boundaries rather than universal ranges.

Study/context	Configuration and scale	Metric or finding	Market/policy boundary	Agricultural boundary	Transferable interpretation
Germany [43]	Elevated pilot and modeled scale-up.	Higher structure cost evaluated against combined performance.	German tariffs, policy, engineering and financing assumptions.	Continued arable production.	Policy and design can offset cost, but result is not a generic CAPEX premium.
Niger [46]	Modeled food-energy case.	Economic feasibility assessed in a land- and resource-constrained setting.	Local energy access, prices and financing.	Crop income and food-energy context.	AV value can be high where energy access and land co-use dominate conventional LCOE comparisons.
Europe [47,48]	Geospatial/configuration comparisons.	Capacity density and design strongly alter technical-economic potential.	European land, irradiance and cost assumptions.	Arable-land suitability and access.	Regional potentials require archetype-specific density and eligibility.
Sweden/northern systems [26,42]	Experimental/model validation.	Economics tied to vertical geometry, northern irradiance and local agriculture.	Swedish prices and seasonal generation.	Northern crop calendar and machinery.	High-latitude evidence is valuable but should not be transferred to southern markets unchanged.
Sweden, 47 ha vertical case [49]	Farmer- and investor-led business models.	Reported farmer-led NPV ≈ EUR 5.5 million, ROI 8.3%, discounted payback 15.1 years.	Case-specific electricity price, discount rate, scale and contracts.	Crop profitability and value allocation differ by ownership model.	Profitability and equity can diverge; the business model is part of system design.
Systematic evidence [50]	Multiple crops, countries and configurations.	Productivity and profitability vary widely across assumptions.	Heterogeneous incentives and markets.	Different yield baselines and crop values.	Report distributions and sensitivities; do not present one payback benchmark.

5.4. Scale, Replicability, and Risk

Scalability requires modularity in both engineering and contracts. Conventional solar farms benefit from standardized piles, racking, electrical blocks, construction sequences, and operating procedures. AV adds crop-specific height, spacing, access, irrigation, and liability requirements. A scalable market will therefore require a limited family of validated archetypes rather than a unique design for every field.

Geospatial assessments can identify theoretical potential, but they should screen slope, field size, crop, protected land, drainage, land tenure, proximity to substations, feeder capacity, road access, and agricultural machinery. Brazilian potential analysis illustrates how regional land and system conditions change feasibility [51], while comparative European configuration research shows that layout assumptions materially affect energy estimates [48].

Operational risk remains under-characterized. Long-term studies should record module degradation, corrosion, cable and connector exposure to agricultural operations, tracking downtime, foundation movement, crop damage, soil compaction, insurance claims, and changes in farming practice. Aquavoltaic and agricultural comparisons also show that “dual use” covers very different operational environments and should not be treated as one risk class [52].

5.5. Policy-Driven Economics

Tariffs, agricultural subsidies, tax classification, permitting time, grid charges, and curtailment rules can move a project from viable to non-viable without changing the hardware. Market-transformation research argues that AV adoption requires policy that recognizes the additional services and costs of dual use [53]. Legal analysis likewise shows that uncertainty over land status and agricultural eligibility creates transaction cost and financing risk [54]. The economic model should therefore include policy variables explicitly rather than treat incentives as an afterthought.

6. Policy, Legal, Social, and Equity Conditions for Deployment

The most persistent deployment barrier is institutional fit. Energy law, agricultural support, land-use planning, and grid regulation were generally designed for separate sectors. AV succeeds when these systems recognize a joint asset and define how agricultural continuity will be measured. It fails when projects can access energy value only by losing agricultural status, or retain agricultural status without demonstrating meaningful farming.

6.1. Comparative Policy Models

Policy pathways differ by governance scale. Germany and Japan have used agricultural-continuity requirements and legal clarification to distinguish dual-use projects from conventional solar development [43,55]. The European policy literature emphasizes harmonized definitions, subsidy compatibility, and planning readiness [53,54,56,57]. The Joint Research Centre identifies the absence of a common EU definition and the resulting risk to agricultural subsidy eligibility as a major obstacle [41].

The United States relies more heavily on state policy, research programs, and project-specific land-use arrangements. The U.S. Department of Energy FARMS program funded impact studies, scalable resources, and models intended to lower barriers for farmers and rural communities [58]. Evidence from Massachusetts shows that policy can enable collaboration and farmer participation while simultaneously creating operational, liability, and data burdens [59].

China supports “agriculture-PV complementarity” and other PV-plus-agriculture models within rural revitalization and renewable-energy policy. National guidance recognizes land-efficiency and agricultural-development benefits, while implementation and land-use supervision remain distributed across ministries and local authorities [60]. India’s PM-KUSUM is not an AV standard, but its Component A supports decentralized ground- or stilt-mounted 0.5–2 MW projects by farmers, farmer groups, cooperatives, panchayats, and producer organizations, creating an institutional route through which dual-use designs may be deployed [61]. India-focused research nevertheless stresses the need for site suitability, water-energy-food integration, finance, and credible agricultural safeguards [44,62].

Table 4. Comparative policy models, safeguards, and failure risks in major agrivoltaic markets.

Region/model	Primary mechanism	Agricultural safeguard	Enabling feature	Failure risk	Transferable lesson
EU/member states [41,53,54,56,57]	Planning rules, agricultural status, national incentives, EU agricultural policy.	Definitions and productivity/continuity requirements vary.	Large quantified land potential and policy recognition.	Fragmentation, subsidy ineligibility, inconsistent permits.	Harmonize minimum definitions while allowing crop- and climate-specific performance thresholds.
United States [58,59,63]	State siting/tariff rules plus federal R&D and demonstration.	Project- or state-specific agricultural plans and monitoring.	Experimental diversity and stakeholder-led learning.	Patchwork rules, liability, data burdens and uneven grid access.	Use research funding to build standards, but connect pilots to durable permitting and finance.
China [57,60]	National support for PV-agriculture integration with local implementation.	Land-use and agricultural requirements administered across authorities.	Rapid deployment capacity and rural-development linkage.	Nominal farming, inconsistent local enforcement, land conversion concerns.	Pair scale with auditable agricultural outcomes and transparent local supervision.
India [44,61,62]	Farmer-oriented decentralized solar and capital/contract support.	Stilted/ground designs can preserve farming; AV-specific safeguards vary.	Farmer groups and cooperatives are eligible project actors.	Finance, water use, grid constraints and unequal contracting.	Build AV criteria into existing farmer-energy programs rather than assume stilt mounting guarantees dual use.
Japan/Germany [43,55]	Solar-sharing or dual-use recognition tied to agricultural activity.	Minimum agricultural continuity/productivity expectations.	Clearer legal identity for dual-use land.	Monitoring burden and incentives to satisfy only minimum thresholds.	Define measurable outcomes, review them periodically, and prevent agricultural tokenism.

Table 4. Comparative policy models, safeguards, and failure risks in major agrivoltaic markets.

Region/model	Primary mechanism	Agricultural safeguard	Enabling feature	Failure risk	Transferable lesson
EU/member states [41,53,54,56,57]	Planning rules, agricultural status, national incentives, EU agricultural policy.	Definitions and productivity/continuity requirements vary.	Large quantified land potential and policy recognition.	Fragmentation, subsidy ineligibility, inconsistent permits.	Harmonize minimum definitions while allowing crop- and climate-specific performance thresholds.
United States [58,59,63]	State siting/tariff rules plus federal R&D and demonstration.	Project- or state-specific agricultural plans and monitoring.	Experimental diversity and stakeholder-led learning.	Patchwork rules, liability, data burdens and uneven grid access.	Use research funding to build standards, but connect pilots to durable permitting and finance.
China [57,60]	National support for PV-agriculture integration with local implementation.	Land-use and agricultural requirements administered across authorities.	Rapid deployment capacity and rural-development linkage.	Nominal farming, inconsistent local enforcement, land conversion concerns.	Pair scale with auditable agricultural outcomes and transparent local supervision.
India [44,61,62]	Farmer-oriented decentralized solar and capital/contract support.	Stilted/ground designs can preserve farming; AV-specific safeguards vary.	Farmer groups and cooperatives are eligible project actors.	Finance, water use, grid constraints and unequal contracting.	Build AV criteria into existing farmer-energy programs rather than assume stilt mounting guarantees dual use.
Japan/Germany [43,55]	Solar-sharing or dual-use recognition tied to agricultural activity.	Minimum agricultural continuity/productivity expectations.	Clearer legal identity for dual-use land.	Monitoring burden and incentives to satisfy only minimum thresholds.	Define measurable outcomes, review them periodically, and prevent agricultural tokenism.

6.2. Social Acceptance and Co-Design

Acceptance depends on landscape change, trust, perceived agricultural authenticity, local benefit, and participation. French case analysis shows that the label “agrivoltaic” does not by itself reconcile energy and agriculture [15]. Survey and stakeholder research identifies concerns about visual impact, land control, food production, and developer motives [63,64,65]. Projects are more likely to gain legitimacy when farmers participate in layout, crop planning, monitoring, and revenue decisions rather than serve only as land lessors.

Co-design should begin before the array geometry is fixed. Machinery routes, crop rotation, drainage, spraying, livestock safety, emergency access, and future farm succession can materially alter layout. Participation is not only a communication strategy; it is a source of engineering requirements. Transparent reporting of agricultural outcomes and grievance mechanisms can also reduce the information asymmetry between developers, authorities, and local communities.

6.3. Equity Defects and Ownership Models

Equity problems arise when policies reward installed capacity but do not specify who receives value or bears operational risk. Common defects include incentives captured mainly by developers or landowners; exclusion of tenant farmers and farm labor from decision-making; up-front capital and credit barriers; weak access to interconnection; long contracts that constrain future farming; and liability or data requirements that smallholders cannot absorb. Equity and justice scholarship therefore treats ownership, procedural participation, and distribution as core deployment variables [66,67,68].

The distinction between landowner and active farmer is especially important. A lease payment can benefit an owner while reducing flexibility or income for a tenant operator. Similarly, an investor-led project may be financially successful while transferring agricultural, curtailment, or maintenance risk to the farmer. Cooperative and community-energy models can broaden participation, but they require governance capability, finance, and technical support [69,70,71].

Table 5. Ownership models and their principal equity implications.

Ownership model	Value pathway	Main advantage	Core equity risk	Recommended safeguard
Developer lease	Fixed or indexed land payment; developer owns energy asset.	Low capital requirement for landowner.	Tenant/farm-worker exclusion; long-term loss of control; agricultural risk shifted outward.	Separate consent and benefit provisions for active farmers; enforceable farming plan.
Farmer-owned	Farmer retains electricity and crop value.	Strong alignment between design and farm operations.	Capital, credit, technical capacity and concentration of downside risk.	Concessional finance, extension support, insurance and standardized contracts.
Joint venture/revenue share	Energy and agricultural value shared by negotiated formula.	Combines developer capital with farm knowledge.	Asymmetric information, opaque deductions, and imbalance in bargaining power.	Transparent cash-flow waterfall, independent advice, audit rights and risk allocation.
Cooperative/community	Collective ownership or local equity participation.	Broader local value retention and legitimacy.	Governance complexity and unequal participation within the community.	Inclusive membership, voting rules, professional management and benefit reporting.
Public/extension-backed pilot	Research or public funding reduces early risk.	Generates open evidence and supports underserved farmers.	Pilot dependence and weak transition to commercial finance.	Predefined scale-up pathway, open data, procurement standards and post-grant support.

Table 5. Ownership models and their principal equity implications.

Ownership model	Value pathway	Main advantage	Core equity risk	Recommended safeguard
Developer lease	Fixed or indexed land payment; developer owns energy asset.	Low capital requirement for landowner.	Tenant/farm-worker exclusion; long-term loss of control; agricultural risk shifted outward.	Separate consent and benefit provisions for active farmers; enforceable farming plan.
Farmer-owned	Farmer retains electricity and crop value.	Strong alignment between design and farm operations.	Capital, credit, technical capacity and concentration of downside risk.	Concessional finance, extension support, insurance and standardized contracts.
Joint venture/revenue share	Energy and agricultural value shared by negotiated formula.	Combines developer capital with farm knowledge.	Asymmetric information, opaque deductions, and imbalance in bargaining power.	Transparent cash-flow waterfall, independent advice, audit rights and risk allocation.
Cooperative/community	Collective ownership or local equity participation.	Broader local value retention and legitimacy.	Governance complexity and unequal participation within the community.	Inclusive membership, voting rules, professional management and benefit reporting.
Public/extension-backed pilot	Research or public funding reduces early risk.	Generates open evidence and supports underserved farmers.	Pilot dependence and weak transition to commercial finance.	Predefined scale-up pathway, open data, procurement standards and post-grant support.

6.4. Policy Design Principles

Five design principles emerge. First, define AV as continued, measurable agricultural production combined with PV, not merely PV located on agricultural land. Second, align land classification, agricultural payments, energy incentives, and taxation so that genuine dual use is not penalized. Third, require a baseline agricultural plan and periodic monitoring while allowing climate- and crop-adjusted thresholds. Fourth, make contracts, ownership, grid access, and finance accessible to small and tenant farmers. Fifth, publish comparable energy and agricultural data so that public support creates transferable knowledge rather than isolated demonstrations.

These principles also clarify unsuccessful policy patterns: definitions based only on structure height; incentives paid without outcome monitoring; blanket prohibitions on PV in agricultural zones; and compliance systems so burdensome that only large developers can participate. Successful policy is neither unrestricted deployment nor rigid one-size-fits-all regulation. It combines a clear dual-use identity with measurable outcomes and proportionate administration.

7. Discussion and Future Research

7.1. Critical Cross-Domain Synthesis

The evidence does not support a single “best” AV configuration. Elevated systems are strongest where machinery access, crop value, and shading benefits justify structural cost. Spaced rows are attractive where conventional construction and operational simplicity dominate. Vertical bifacial systems are strongest where the hourly generation profile, rear-side irradiance, northern-season performance, and low overhead obstruction have value. The apparent disagreement among studies often reflects a change in objective or comparator rather than an engineering contradiction.

A second synthesis finding is that performance metrics must be layered. PV yield per installed kilowatt evaluates the generator; PV yield per hectare evaluates land intensity; crop response evaluates agricultural continuity; LER evaluates combined normalized productivity; and NPV or social value evaluates economic allocation. None can substitute for all the others. A credible AV claim should identify which layer it addresses and which it does not.

Third, policy and ownership are endogenous to technical performance. Agricultural monitoring affects operating cost; subsidy status affects crop choice; a lease can constrain layout changes; and interconnection conditions affect the value of a vertical generation profile. Treating governance as a final “barriers” section therefore understates its role. In scaled AV deployment, the engineered system includes the contract, monitoring rule, and land classification as well as the module and structure.

7.2. Limitations of the Evidence and of This Review

The evidence base remains heterogeneous and geographically uneven. Europe, the United States, Japan, China, and India receive substantial attention, while many tropical and low-income farming systems are represented by fewer or smaller studies. Field experiments commonly cover one or a few seasons, whereas PV assets operate for 20–30 years. Crop rotation, soil health, structural degradation, changing electricity markets, and farm succession are therefore poorly observed over a full asset life.

Economic studies frequently use modeled prices, simplified crop revenues, and different definitions of land and project area. Policy changes quickly and may be implemented differently below the national level. Social studies often involve early adopters or hypothetical projects. These limitations justify the boundary-conditioned synthesis used here and preclude a pooled LER, crop-yield effect, CAPEX premium, or payback period.

This review also has methodological limits. The original submission did not retain database hit counts, and the revised work is therefore an integrative critical review rather than a PRISMA systematic review. Although the search logic, eligibility criteria, and appraisal framework are now explicit, English-language selection and publisher-access pathways may have omitted relevant regional or non-English evidence. The appraisal is structured but not a domain-specific risk-of-bias instrument for every disciplinary method. Conclusions should therefore be read as a transparent synthesis of the cited corpus, not as an exhaustive census of all AV publications.

7.3. Research Roadmap for Energy-System Integration

The core next step is to move from isolated project metrics to model-ready AV archetypes. The following roadmap links field measurement, engineering models, spatial eligibility, capacity expansion, and policy validation.

Table 6. Staged research roadmap for incorporating agrivoltaics into solar-energy and capacity-expansion models.

Stage	Required work	Minimum data/output	Modeling use	Validation criterion
1. Standardized field data	Collect paired PV, PAR, crop, microclimate, water, geometry, cost and operational data.	Hourly generation; front/rear irradiance; crop yield/quality; weather; row geometry; land area; costs.	Creates comparable training and validation datasets.	Matched agricultural and PV references over multiple seasons.
2. Configuration archetypes	Parameterize elevated-fixed, elevated-tracking, spaced-tilted, vertical-bifacial and greenhouse classes.	CAPEX/OPEX, capacity density, hourly profile, crop response functions, operational constraints.	Technology choices in planning models.	Out-of-sample validation across sites and crops.
3. Spatial eligibility	Overlay crop, climate, slope, biodiversity, tenure, machinery, water, road and grid constraints.	Eligible land by archetype and adoption scenario.	GIS supply curves and regional potentials.	Comparison with permitting outcomes and farmer adoption.
4. Capacity expansion	Add AV as a multi-output technology with agricultural constraints and temporal generation.	Cost and performance distributions; grid connection; crop-retention floor; scenario weights.	Least-cost, multi-objective or stochastic energy pathways.	Benchmark against dedicated PV plus separate agriculture.
5. Policy/ownership scenarios	Test definitions, incentives, monitoring costs, finance and ownership distributions.	Stakeholder cash flows, benefit incidence, compliance cost and adoption response.	Policy robustness and just-transition analysis.	Observed contracts, project continuation and distributional outcomes.
6. Long-term feedback	Update archetypes with degradation, maintenance, crop rotation, climate extremes and market change.	Longitudinal reliability and economic data.	Adaptive planning and bankability.	20–30-year digital records or validated degradation/transition models.

Table 6. Staged research roadmap for incorporating agrivoltaics into solar-energy and capacity-expansion models.

Stage	Required work	Minimum data/output	Modeling use	Validation criterion
1. Standardized field data	Collect paired PV, PAR, crop, microclimate, water, geometry, cost and operational data.	Hourly generation; front/rear irradiance; crop yield/quality; weather; row geometry; land area; costs.	Creates comparable training and validation datasets.	Matched agricultural and PV references over multiple seasons.
2. Configuration archetypes	Parameterize elevated-fixed, elevated-tracking, spaced-tilted, vertical-bifacial and greenhouse classes.	CAPEX/OPEX, capacity density, hourly profile, crop response functions, operational constraints.	Technology choices in planning models.	Out-of-sample validation across sites and crops.
3. Spatial eligibility	Overlay crop, climate, slope, biodiversity, tenure, machinery, water, road and grid constraints.	Eligible land by archetype and adoption scenario.	GIS supply curves and regional potentials.	Comparison with permitting outcomes and farmer adoption.
4. Capacity expansion	Add AV as a multi-output technology with agricultural constraints and temporal generation.	Cost and performance distributions; grid connection; crop-retention floor; scenario weights.	Least-cost, multi-objective or stochastic energy pathways.	Benchmark against dedicated PV plus separate agriculture.
5. Policy/ownership scenarios	Test definitions, incentives, monitoring costs, finance and ownership distributions.	Stakeholder cash flows, benefit incidence, compliance cost and adoption response.	Policy robustness and just-transition analysis.	Observed contracts, project continuation and distributional outcomes.
6. Long-term feedback	Update archetypes with degradation, maintenance, crop rotation, climate extremes and market change.	Longitudinal reliability and economic data.	Adaptive planning and bankability.	20–30-year digital records or validated degradation/transition models.

7.4. Implications for Solar Research and Practice

For the solar research community, AV should be evaluated as a family of PV system architectures with non-electrical operating constraints. Module and array research can contribute through bifacial modeling, spectral and albedo measurement, dynamic tracking control, reliability, lightweight structures, and power-electronic designs that reduce mismatch. Solar-resource and grid studies can quantify the temporal value of vertical and tracking profiles. Techno-economic work should report stakeholder-specific cash flows rather than only project-level profitability.

For practitioners, the immediate implication is to specify the project boundary before presenting performance. Every design proposal should state configuration, height, row spacing, ground coverage ratio, module technology, crop, machinery, reference system, project area, electricity market, financing, agricultural revenue, and policy eligibility. Without these variables, a numerical LER, cost premium, or payback period is not sufficiently interpretable for investment or planning.

8. Conclusions

Agrivoltaics can reduce conflict between photovoltaic expansion and agricultural land use, but its value is configuration- and context-dependent. Elevated and tracking systems offer agricultural access and controllable shade at higher structural and operational cost. Spaced systems are simpler but reduce PV density. Vertical bifacial systems can provide accessible fields and distinctive generation profiles, while their output depends strongly on row spacing, latitude, rear irradiance, albedo, and electrical design. No configuration should be promoted without an explicit comparator and project boundary.

Land equivalent ratios above unity demonstrate the possibility of joint land-productivity gains, not a universal performance guarantee. LER should be decomposed into crop and electricity components and supplemented with energy per hectare, hourly generation, crop quality, and uncertainty. For energy-transition analysis, AV must be represented through configuration-specific capacity density, eligible land, hourly profiles, cost distributions, and agricultural-performance constraints. The EU potential example demonstrates that even lower-density dual-use PV can materially affect capacity pathways when large agricultural land areas are considered, but technical potential must be reduced by legal, grid, ecological, ownership, and adoption constraints.

Economic and policy conclusions are equally conditional. Cost and payback vary with structure, scale, crop, electricity price, financing, incentives, and ownership. Scalable deployment therefore requires standardized engineering archetypes, measurable agricultural safeguards, aligned energy and agricultural policy, transparent contracts, and finance that enables active farmers not only landowners and developers to participate. Long-term paired datasets and model-ready archetypes are the highest-priority research outputs for establishing agrivoltaics as credible, bankable, and equitable solar-energy infrastructure.