Should Agrivoltaics Ever Be Decommissioned? How Agrivoltaics Bolster Farm Climate Adaptation Even When Unpowered

Abstract

Solar photovoltaic systems now produce the lowest-cost electricity in history and coupling with agriculture in agrivoltaics increases crop yields. This indicates solar will continue to experience explosive growth. Concerns exist, however, about the long-term end-of-life decommissioning of solar farms. For example, due to fossil fuel decommissioning mismanagement, Alberta is inundated with orphaned oil and gas wells that have remediation cost estimates of CAD$100 billion. Such comparisons have prompted preemptive legislation targeting solar farms, but is the fear justified? This study addresses this question by (1) analyzing warranted and actual lifespans of key agrivoltaic system components, (2) experimentally measuring microclimate impacts of two agrivoltaic arrays (fully powered with electricity extraction and unpowered to simulate post-inverter-failure conditions) and (3) quantifying agrivoltaic yield gains based on crops previously shown to respond positively to such conditions. Experimental results indicate that unpowered photovoltaic shading not only moderates soil temperatures but also enhances soil moisture conservation relative to unshaded conditions. This study demonstrates that agrivoltaic systems, even after the cessation of power generation, can continue to deliver meaningful agronomic and economic value through passive shading and policy frameworks should adapt to this dual-use reality. Integrating agronomic co-benefits into decommissioning policy supports long-term farm productivity and climate resilience.

1. Introduction

The acceleration of anthropogenic climate change [1], primarily driven by carbon dioxide emissions from fossil fuel combustion [2], poses a significant threat to global food systems [3,4]. Rising average temperatures, along with more frequent and intense heatwaves, are expected to substantially reduce crop yields in the coming decades [5]. As global temperatures increase, so too does the necessity of climate-resilient agricultural strategies [6]. Among the adaptive techniques being explored, passive shading of crops has emerged as a viable means to mitigate heat stress [7]. These solutions, however, typically entail ongoing costs without providing any direct co-benefits that create additional value such as energy production. Agrivoltaics—the dual use of agricultural land for both crop production and solar energy generation—offers a compelling solution by integrating crop shading with the economic benefits of solar photovoltaic (PV) technologies [8]. Farmers can either own and operate agrivoltaic systems to sell electricity to the grid or benefit from lease agreements with third-party solar developers. As a result, agrivoltaics represents a self-financing shading system that can improve crop performance [9,10,11,12] while diversifying farm income [13]. Globally, agrivoltaic adoption is accelerating, with projects deployed across Europe [14], Asia [15,16,17], and North America [18] in response to both land use efficiency and climate adaptation imperatives [19,20].

Concerns, however, have been raised about the long-term environmental risks of solar installations, particularly with regard to end-of-life decommissioning [21,22,23]. Alberta, for instance, has experienced a proliferation of orphaned oil and gas wells, with the cost of remediation projected to exceed CAD $100 billion [24]. Such comparisons have prompted preemptive legislation targeting solar farms, raising questions about long-term land stewardship in the renewable energy sector and PV in particular [25].

Given these concerns, it is essential to evaluate whether agrivoltaic systems—especially at the end of their electrical service life—continue to offer agronomic benefits that justify their physical presence on farmland. This study addresses this question by conducting a multi-phase investigation. First, the warranted and actual lifespans of key agrivoltaic system components, distinguishing between mechanical (e.g., racking, footings and PV modules) and electrical (e.g., PV modules energy generation and inverters) elements were evaluated. Second, the microclimate impacts of two agrivoltaic arrays: one fully powered with electricity extraction, and the other unpowered to simulate post-inverter-failure conditions are compared. Using in situ measurements of air and soil temperature as well as soil humidity, shading benefits in unpowered systems were assessed. Third, the yield gains associated with agrivoltaic shading based on crops previously shown to respond positively to such conditions were quantified in both agronomic and economic terms. Finally, how these benefits might change under projected future climate conditions, to assess whether unpowered agrivoltaics can be considered a semi-permanent climate adaptation strategy. Unpowered agrivoltaics are defined as PV arrays that are no longer connected to an inverter or a load (e.g., they are not producing electricity), but that continue to provide shading and microclimatic modifications through their physical structure on farmland. This scenario can occur after inverter failure, at the end of a project’s financial lifetime, or if arrays are intentionally repurposed for passive agricultural benefits.

2. Materials and Methods

2.1. Component Lifetimes

To assess the long-term value of agrivoltaic systems, published warranties and field data on the operational lifespans of key components was reviewed. Electrical components, particularly inverters, typically have the shortest expected lifespans [26]. Photovoltaic modules, by contrast, can continue to function for longer—albeit with reduced output; for example, crystalline silicon modules typically exhibit a reduction in power output of approximately 0.5% per year [27,28,29,30]. Purely mechanical components, including racking systems and concrete or screw footings, generally have even longer operational lifespans [31]. For instance, PV racking structures in long-standing installations in Europe remain operational after more than 35 years of field exposure, with refurbishments focused primarily on electrical components rather than structural systems [32]. The implication is that the shading functionality of PV modules and their structural supports may outlast their ability to generate or deliver electricity, presenting an opportunity to evaluate their passive value to agriculture in an unpowered state.

2.2. Experimental Design

Two similar agrivoltaic arrays were installed on test plots, each comprising a series of fixed-tilt, bifacial crystalline silicon (c-Si) PV modules mounted on stilt-mounted racking [33]. The treatments included modules with a nominal light transmittance of 8% [34] and 44% [35], manufactured by Brite Solar, Thessaloniki, Greece. The powered system was installed at a farm in Denfield, Ontario, Canada, while the unpowered system was deployed at the Western Innovation for Renewable Energy Deployment (WIRED) facility [36], located at the Western University Field Station in Ilderton, Ontario, Canada, approximately 14 km from the Denfield site (Figure 1). This design choice was necessitated by infrastructure availability and logistical constraints. As a result, background conditions such as soil type, drainage, and local microclimate may have differed between sites. While both sites are located in the same agro-climatic zone and subject to similar weather conditions, the possibility of confounding site effects should be considered when interpreting differences between powered and unpowered treatments. Vegetation differences, for example, grass vs. bare soil, may have influenced temperature dynamics. To ensure methodological rigor, each system was benchmarked against its respective control plot, enabling like-to-like comparisons. At the powered site (Denfield), the control plot consisted of bare-soil conditions, while at the unpowered site (Ilderton), the control consisted of open farmland under grass cover managed as pasture. These control plots enabled direct comparisons of shaded versus unshaded conditions at each location. Thus, the experiment comprised four treatment types in total: powered 44% c-Si array vs. grassed control (Denfield) and unpowered 8% and 44% c-Si arrays vs. bare-soil control (Ilderton).

In the first array (Denfield), electricity was extracted to a grid-tied system, maintaining the system in a powered state where the fraction of the energy converted to electricity striking the modules was removed from the local environment. In contrast, the second array (Ilderton) was configured to mimic an unpowered condition by disconnecting electrical outputs, simulating a post-inverter/PV module-failure scenario. In the unpowered array, absorbed solar energy is predominantly dissipated as heat due to carrier recombination, whereas in the powered array, a portion of this energy is exported as electricity. A complete surface-energy-balance analysis would be required to quantify these fluxes, which was beyond the scope of this study (for example, see Mattei et al. [37] and Armstrong et al. [38]). In both arrays, soil temperature measurements were collected at fixed intervals during peak summer conditions, with results compared to their respective unshaded control plots.

To capture additional microclimatic impacts of agrivoltaics, air temperature measurements were recorded for the 44% c-Si treatment and its control in the powered system, as this site aimed to evaluate potential thermal effects on crop canopies from active electricity generation. Conversely, soil humidity measurements were conducted at the unpowered site, where shading effects were expected to influence soil water retention more strongly in the absence of electrical operation. This complementary sensor deployment enabled site-specific insights into how powered and unpowered agrivoltaic systems differentially influence above- and below-ground microclimatic conditions.

Temperature and soil moisture were measured using Jericho TH-1 and Teros 10 by Meter Group, respectively. Each probe was installed at a depth of 7.6 cm (3 inches). For the probes, careful calibration was used to ensure accuracy. A three-point calibration was conducted in cold, warm, and hot water baths, yielding accuracies of ±0.5 °C for temperature, while the soil humidity sensor was factory calibrated at ±0.03 m³/m³ for volumetric water content. Logging was performed at 1 s intervals and down sampled to 5 min and 15 min for unpowered and powered arrays. Quality control included visual inspection of time series data, removal of spurious spikes, and consistency checks across diurnal cycles. Given the rigorous calibration and stable performance of the probes, single-sensor deployment was deemed sufficient for this pilot-scale study.

2.3. Statistical Analysis

Raw data were screened for quality assurance, and any incomplete or erroneous values (e.g., sensor dropouts, outliers) were excluded. The dataset was then compiled into treatment-wise time series for analysis. Daily averages were calculated for each treatment by generating pivot tables in a spreadsheet, grouping all sensor readings by date and treatment type.

For each variable (soil temperature, air temperature, and soil humidity), descriptive statistics were calculated, including mean, standard deviation (SD), standard error (SE), variance, skewness, and kurtosis. Treatment differences were assessed using one-way Analysis of Variance (ANOVA) (unpowered array: 8% c-Si, 44% c-Si, control; powered array: 44% c-Si vs. control). ANOVA outputs are reported with the F statistic, degrees of freedom, and associated p-values, alongside the critical F value for the given α = 0.05 significance threshold.

2.4. Yield-Responsive Crops

Data from published agrivoltaic studies identifying crops that exhibit increased yield or quality under partial shading conditions was compiled.

Table 1. Increased crop yields under agrivoltaics from past literature *.

Country	Low	High	Transparency (%)	Sources
Germany	2.7% (winter wheat)	11% (potatoes)	64.9%	[39]
	3% (winter wheat)	12% (celeriac)	70.0%	[40]
	31.9% (Celeriac)	48% (Celeriac)	70.0%	[41]
Italy	-	69% (chicory)	100.0% (Plants under the shadows of modules)	[42]
U.S.	200% (tomatoes)	300% (chiltepin pepper)	38.4%	[43]
	-	90% (pasture)	50.0%	[44]
China	-	23% (Jerusalem artichoke)	33.3%	[45]
Japan	-	5.5% (corn)	55.3%	[46]
Canada	4% (strawberries)	18% (strawberries)	44–69%	[10]
Malysia	-	159% (Andrographis paniculata)	59.08%	[47]
	-	2.9% (okra)	38.5%	[48]
	-	2% (Chinese cabbage)		[48]
	-	37.5% (pennywort)		[48]

* Adopted from Agrivoltaics as a Systems Innovation: Multi-Dimensional Benefits from Global Studies Across Climate, Agriculture, Energy, and Ecosystems (To be published). As the supporting manuscript remains under review, these tables should be regarded as provisional.

reports the ranges of crop yield increase under agrivoltaic systems across various countries and crop types. Yield improvements are expressed in percentages for diverse crops, including cereals, vegetables, and herbs. The analysis is restricted to a fixed-tilt array with c-Si PV module technology. Reported transparencies are approximate, derived from module width and inter-row spacing, and thus, they provide non-uniform shading. It is important to note that in this study, the 8% and 44% transparent c-Si modules reflect manufacturer-reported optical transmittance values used in the experiment. These were agrivoltaic-specific bifacial modules, whereas the transparency values presented in Table 1 for the literature synthesis represent effective shading fractions derived from conventional (fully opaque) modules, and the semi-transparency comes from module width and inter-row spacing, leaving unshaded gaps.

Table 2. Crop types, national and global production volumes (in million tonnes), and market prices for selected crop-country combinations *.

Country	Crop	Country Production (Million Tonnes)	Price (Per tonne)	World Production (Million Tonnes)
Germany	Winter wheat	21.8 [49]	179.6 [50]	808.521 [51]
	Potatoes	11.6 [52]	3080 [53]	375 [54]
	Celeriac	0.0845 [55]	1540 [56]	-
Italy	Chicory	0.0006382 [57]	2630 [58]	0.032662 [59]
U.S.	Tomatoes	16 [60]	2000 [61]	170 [60]
China	Jerusalem artichoke	0.0007436 [62]	5400 [63]	75 [64]
Japan	Corn	0.016 [65]	240 [66]	1220 [67]
Canada	Strawberries	0.02349 [68]	6510 [10]	8.885028 [69]
Malaysia	Okra	0.063074 [70]	320 [71]	11.232656 [72]
	Chinese cabbage	0.15 [73]	1470 [74]	0.048241 [75]

* Adopted from Agrivoltaics as a Systems Innovation: Multi-Dimensional Benefits from Global Studies Across Climate, Agriculture, Energy, and Ecosystems (To be published). As the supporting manuscript remains under review, these tables should be regarded as provisional.

outlines the crop type for each crop-country pairing, along with national and global production volumes (measured in million tonnes) and associated market prices. These data were compiled from various publicly accessible agricultural databases and commodity pricing sources.

Market-based examples of agrivoltaic systems further corroborate these findings. In France, the Sun’Agri pilot program reported grape yields 20–60% higher under dynamic PV shading compared to open-field controls [76]. Similarly, agrivoltaic trials in Granges-sur-Lot demonstrated up to 50% higher eggplant harvests under modules than in unshaded plots [77]. In addition, Ombrea’s asparagus demonstration site in Lot-et-Garonne revealed improved soil temperature regulation and crop resilience under PV cover [78]. These market-linked case studies demonstrate that shading benefits are not only theoretical but have already translated into measurable agricultural and economic gains.

3. Results

3.1. Component Lifespan Summary

Figure 2 presents the range of lifetimes for key agrivoltaic system components. The inverter, as expected, demonstrates the shortest lifespan. PV modules and racking systems, however, remain structurally viable for shading even after their power-generating function diminishes or ceases. Based on this analysis, PV systems can continue to provide agronomic value in the form of passive shading for almost 100 years beyond inverter failure (

Table 3. Operational lifespan of agrivoltaic system components.

Component	Expected Functional Lifespan Range	Key Notes and Sources
PV module (electrical)	25–35 years for electrical operation	- Monocrystalline Si: 25–30 year typical [81], up to 35 yr [82] - Polycrystalline Si: 20–25 yr [82] - Thin-film: 15–20 yr [82] - Warranties guarantee ≥ 80% of rated power at year 25 [83].
PV module (mechanical/glass)	35–1000 years	- Simulated medieval glasses showed durability around 103 years (approximately one millennium), while highly stable glass types may last up to 106 years [84] - Advanced encapsulants (e.g., POE, silicone-based) have demonstrated projected service lifetimes of >35 years [85]
Inverter	5–20 years	- 5–10 yr, extendable to 20 yr [86]
Racking structure (stainless teel)	40–100 years	Stainless-steel systems routinely last the life of the PV array [87,88,89]
Foundations and footings	40–200 years	- Hot dip galvanized steel piles can last up to 40 yr [90] - Concrete foundations can last up to 200 years [91]

). While tempered glass used in PV modules possesses the intrinsic durability to last for centuries, actual operational lifespan is typically constrained by encapsulant degradation—particularly delamination and adhesive failure [79,80].

3.2. Microclimate Comparison: Powered vs. Unpowered

Significant soil temperature reductions were recorded beneath the 44% transparent PV modules in both powered and unpowered conditions (Figure 3). For the powered array, the mean soil temperature under the 44%-transparency modules was 20.7 °C, compared to 21.7 °C in the unshaded control. This represents an average reduction of 1.0 °C over the two-month measurement period, with a maximum instantaneous reduction of 5.0 °C. In addition, air temperature measurements indicated a cooling effect under the 44%-transparent c-Si modules, with a mean temperature of 21.8 °C, approximately 0.3 °C lower than the control (22.1 °C), and a maximum reduction of 6.8 °C during peak daytime conditions.

For the unpowered array, shading effects were more pronounced (Figure 4). The mean soil temperature under the 8% transparent modules was 21.3 °C, and under the 44% transparent modules 20.7 °C, compared to 22.2 °C in the control. This equates to an average reduction of 0.9 °C beneath the 8% modules, with a maximum single-point reduction of 14.1 °C; and an average reduction of 1.6 °C beneath the 44% modules, with a maximum single-point reduction of 11.9 °C. Although the magnitude of reduction appeared stronger under the unpowered system, this may partly reflect differences in vegetation and substrate, which can influence soil albedo, evapotranspiration, and heat capacity. Nonetheless, in both systems, the agrivoltaic treatments consistently lowered soil temperatures compared to unshaded plots.

Complementary soil humidity measurements further supported these shading effects. The control plot consistently exhibited lower moisture retention (0.231 m³/m³) compared to the agrivoltaic treatments, which averaged 0.245 m³/m³ beneath the 8% modules, and 0.234 m³/m³ beneath the 44% modules. These results indicate that unpowered PV shading not only moderates soil temperatures, but also enhances soil moisture conservation relative to unshaded conditions. These properties have already been shown to be beneficial for a wide range of crops (Table 1) and are now confirmed for unpowered agrivoltaics.

To further validate these findings, thermal imagery from the unpowered site is included (Figure 5), which visually illustrates the cooler soil surface temperatures beneath agrivoltaic modules compared to adjacent unshaded areas. For example, under the 8% c-Si PV modules, the temperature under the shade is approximately 20 °C as compared to around 40 °C on the outside.

Single-day soil temperature profiles were examined for both the powered and unpowered systems to provide finer resolution of diurnal dynamics (Figure 6). In both the powered and the unpowered system, soil temperatures remained consistently lower beneath the 44%c-Si modules compared to the control.

3.3. Statistical Analysis

For the powered array, soil temperature under the 44% c-Si treatment averaged 20.71 °C, compared to 21.73 °C in the unshaded control. Variability was relatively low across treatments (SD ≈ 1.9 °C for 44% and 1.8 °C for the control). ANOVA indicated a highly significant treatment effect (F(1, 4882) = 359.19, p < 0.001), confirming that the 44% c-Si treatment significantly reduced soil temperatures relative to the control.

Air temperature followed a similar but weaker pattern, averaging 21.81 °C under the 44% c-Si array compared to 22.11 °C in the control (SD ≈ 5.07 °C vs. 5.20 °C, respectively). ANOVA results showed a small but statistically significant difference (F(1, 4882) = 4.22, p = 0.04). This suggests that the shading effect of the 44% c-Si array moderated air temperature compared to the unshaded condition.

For the unpowered array, descriptive statistics indicated that mean soil temperature varied across treatments, with values of 21.31 °C under the 8% c-Si array, 20.68 °C under the 44% c-Si array, and 22.25 °C in the control. Variability was lowest in the 8% treatment (variance = 23.97) and highest in the control (variance = 46.84). The one-way ANOVA showed significant differences in mean temperature among treatments (F = 63.56, df = 2.10218, p < 0.001). Comparisons suggest that both shaded treatments reduced temperature compared to the control, with the 44% array having the greatest cooling effect.

Mean soil humidity was 0.245 m³/m³ under the 8% c-Si array, 0.234 m³/m³ under the 44% c-Si array, and 0.231 m³/m³ in the control. Variability was lowest in the control (variance = 0.00065) and highest in the 44% treatment (variance = 0.00267). ANOVA results indicated significant differences among treatments (F = 95.06, df = 2.10218, p < 0.001). Soil humidity was consistently higher under both shaded treatments compared to the control, with the 8% array producing the greatest increase.

Figure 7 presents the daily temperature patterns for the powered and unpowered arrays, illustrating consistently lower temperatures beneath the modules compared to the respective controls.

3.4. Quantified Yield Gains and Economic Impacts

Yield enhancements were converted into estimated additional farm income by applying current market prices for each crop. As summarized in

Table 4. Indicative estimated additional farm income per year resulting from yield increases under unpowered agrivoltaic shading. Income calculations are based on documented yield improvements and current market prices for each crop *.

Country	Crop	Additional Yield—Country (Million Tonnes)	Additional Income (Million USD)	Additional Yield—World (Million Tonnes)	Additional Income (Million USD)
Germany	Winter Wheat	0.654	117.5	24.3	4356
	Potatoes	1.276	3930.1	41.3	127,050
Italy	Chicory	0.000440344	1.2	0.02	59
U.S.	Tomatoes	16	32,000.0	170.0	340,000
China	Jer. artichoke	0.000171035	0.9	17.3	93,150
Japan	Corn	0.00088	0.2	67.1	16,104
Malaysia	Okra	0.001829146	0.6	0.3	104
	Chinese cabbage	0.003	4.4	0.001	1.4
Canada	Strawberries	0.004228	19.8	1.60	7496

* Adopted from Agrivoltaics as a Systems Innovation: Multi-Dimensional Benefits from Global Studies Across Climate, Agriculture, Energy, and Ecosystems (To be published). As the supporting manuscript remains under review, these tables should be regarded as provisional.

, the results illustrate the potential economic co-benefits of passive shading provided by unpowered agrivoltaic systems. These income calculations are presented as illustrative estimates based on extrapolations from reported crop yield responses under agrivoltaic systems. Given that the underlying studies span diverse contexts, crop management practices, and market conditions, these figures should not be interpreted as precise forecasts, but rather as indicative of the potential economic scale, subject to practical constraints on land use, adoption, and system design.

4. Discussion

4.1. Agronomic Value of Powered Versus Unpowered Agrivoltaics

The experimental results demonstrate that both powered and unpowered agrivoltaic systems can significantly reduce soil temperature relative to unshaded controls. For example, beneath 44% transparent modules, the unpowered system lowered average soil temperature by 1.6 °C compared to the control, with single-point reductions as high as 11.9 °C. In contrast, the powered system produced a smaller average reduction of 1.0 °C and a maximum of 6.8 °C. Although the unpowered site showed more pronounced reductions, the local vegetation and ground cover differences may have amplified this effect. Even accounting for such site-specific variation, the persistence of cooling in both systems supports the broader conclusion that agrivoltaic structures offer enduring microclimate benefits even after electrical generation ceases. Previous studies have shown similar soil temperature reduction characteristics ranging from 1 °C [8] to 5 °C [92].

These findings align with controlled experimental observations that show shaded ground under PV modules can be 3–4 °C cooler, due to reduced solar radiation and increased evapotranspiration effects [93]. Similar results have been reported in agrivoltaic contexts where shading produced soil temperature drops of 0.5–2.3 °C compared to full-sun controls [94]. Cover crops and vegetative covers have been shown to reduce spring time soil temperatures—by intercepting solar radiation and increasing latent heat flux—compared to bare soil, with reductions ranging up to several degrees Celsius; additionally, vegetative covers modify soil thermal properties such as heat capacity and moisture retention, further influencing temperature dynamics [95].

The subsequent implications of reduced soil temperatures are very significant. Studies show that soils with increased temperatures increase the decomposition of soil organic matter [96,97,98,99,100], thus exacerbating soil carbon dioxide efflux and subsequently, contributing to global warming [95]. This is complemented by the fact that previous study on agrivoltaic systems in warmer climates have shown enhanced soil moisture, organic carbon, nitrogen–phosphorus–potassium nutrients, microbial biomass, and urease activity [101].

4.2. Policy Implications

Current policy debates around solar PV decommissioning should be reframed to account for the enduring utility of agrivoltaic structures. While electrical components such as inverters and modules may degrade in electrical performance over time, the mechanical infrastructure retains its function as an agronomic asset through partial shading. The systems can provide shade that can reduce heat stress on crops [102,103], leading to improved growth and yield (Table 1). The shading provided by agrivoltaic systems have also shown to be beneficial for pollination services [42]. In addition, agrivoltaic grazing systems allow for sustainable livestock grazing beneath solar panels by providing cover for animals [104]. Policymakers should consider alternative frameworks for post-electrical-use agrivoltaics—such as permitting reduced regulatory obligations if systems continue to deliver farm productivity benefits. Thus agrivoltaics can be viewed as analogous to shelterbelt programs which serve as effective tools for soil conservation, reducing wind and water erosion, improving soil fertility, and enhancing biodiversity [105]. Similarly agrivoltaics can be treated as shade net installations which also ameliorate solar radiation, temperatures, and relative humidity, leading to improved soil conditions and enhanced vegetable performance [106,107]. Thus, unpowered agrivoltaics represent a passive climate adaptation tool worthy of long-term consideration. Future work is necessary to determine if policies should be adapted to encourage repowering such agronomic assets.

4.3. Study Limitations

While this study provides preliminary evidence that unpowered agrivoltaic systems can continue to deliver valuable agronomic benefits, several limitations should be acknowledged.

4.3.1. Temporal Scope

The duration of the experimental observation—limited to peak summer conditions in one year—may not capture the full variability of plant and soil responses across an entire growing season or multiple climatic conditions. Microclimate effects, particularly those related to soil temperature and humidity, can vary with precipitation patterns [108,109] and local weather fluctuations [110]. As such, short-term measurements may not fully represent long-term shading benefits, particularly in light of the impact of future climate change [111].

4.3.2. Crop Diversity

The study relies on a limited number of crop types and shading configurations. While crops such as lettuce [9,12], strawberries [10,11], and amaranth [19] are well-documented beneficiaries of partial agrivoltaic shading in Canada, the generalizability of results to other crops—particularly grains and root vegetables—remains uncertain. Moreover, the experimental arrays used fixed-tilt stilt mounted [33], bifacial modules [35,112], which may not replicate the shading dynamics of more diverse or advanced agrivoltaic configurations (e.g., tracking systems, or dynamic shading systems).

4.3.3. Structural Integrity

This study assumes that PV modules and racking systems will remain structurally intact and safe after inverter failure. In reality, degradation [113,114,115], wind damage [116], corrosion [117], or vandalism may compromise the physical integrity of modules or mounting systems over time, especially without ongoing maintenance [118]. This raises questions about liability, safety, and the potential for diminishing shading benefits in the absence of electrical function and should be addressed in future work.

4.3.4. Economic Assumptions

The economic valuation of increased crop yields is based on current market prices and yield responses from literature. These values are highly context-dependent and may not reflect the profitability of unpowered agrivoltaics under fluctuating commodity prices [119] or evolving farm input costs [120]. Moreover, the yield-based income values reported are illustrative in nature. They reflect broad extrapolations from diverse agrivoltaic studies and economic assumptions and therefore should not be interpreted as predictive forecasts for specific regions or crops. Instead, these figures serve to demonstrate the potential magnitude of co-benefits, while acknowledging that actual outcomes will depend on adoption ceilings, logistical feasibility, and economic considerations.

4.3.5. Social and Regulatory Factors

The acceptance of leaving unpowered PV arrays on agricultural land as permanent infrastructure may vary depending on local policy frameworks, community attitudes, and perceptions about land stewardship. Although, studies in Canada [121] and the U.S. [122] have shown increased support from agrivoltaics, resistance from regulators or neighbors may limit the practical implementation of the study’s findings, even if agronomic benefits are demonstrable.

4.3.6. Experimental Conditions—Measurement Context

In the unpowered system, temperature was measured in pots surrounded by grass, whereas the powered system measurements were taken with bare soil directly underneath modules. Variations in ground cover and substrate (grass versus soil) are known to significantly impact soil temperature through changes in ground albedo, evapotranspiration, and heat capacity [95]. Research has shown that vegetated surfaces under PV modules can reduce surface temperatures through evapotranspiration compared to bare ground [123]. Therefore, differences in measurement context likely contributed to the observed temperature differentials between powered and unpowered arrays.

4.3.7. Experimental Conditions—Location Context

A key limitation of this study is that powered and unpowered arrays were installed at different sites (Denfield Vs. Ilderton). Although both are within the same regional climate zone, site-level differences in soil composition, moisture retention, and surrounding vegetation may have contributed to microclimate variation. Therefore, the results should not be interpreted as a direct head-to-head comparison, but rather as an indicative demonstration of how powered and unpowered agrivoltaic shading can influence soil temperature and moisture. Future research should aim to evaluate powered and unpowered configurations at the same site under otherwise identical conditions to isolate treatment effects more rigorously.

4.4. Directions of Future Research

Future research should expand on both the temporal and spatial scope of this study to improve the robustness and policy relevance of the findings.

4.4.1. Multi-Season Field Trials

Multi-season and multi-year field trials should be conducted to observe the agronomic effects of unpowered agrivoltaics across entire crop cycles. These studies should include a broader diversity of crops, soil types, and climate zones to enhance generalizability. Crop responses should be measured not only in terms of yield, but also quality, pest and disease incidence, nutrient benefits, and water use efficiency.

4.4.2. Powered Versus Unpowered Comparison

Comparative studies between powered and unpowered systems under identical environmental and agronomic conditions should be extended to quantify the precise biophysical differences—if any—that occur when electrical generation ceases. This includes evaluating potential changes in module surface temperature, longwave radiation absorption by crops, and convective air flows, which may influence plant growth under different energy states of the array.

4.4.3. Mechanical Degradation

Mechanical degradation over time should be studied more rigorously. This includes assessing corrosion of racking systems, structural integrity under snow and wind loads, and the risk of glass delamination or module breakage after electrical failure. Such work would clarify the realistic lifespan of shading functionality, inform maintenance requirements, and help define standards for long-term structural safety in unpowered agrivoltaic systems.

4.4.4. Economic Modeling

Economic models should be developed that incorporate full life-cycle costs and benefits of agrivoltaic installations under various ownership, maintenance, and energy generation scenarios. Sensitivity analyses can help evaluate the robustness of economic outcomes under different assumptions about inverter replacement, land value, crop prices, and climate change impacts. Moreover, open-source, low-cost racking designs [124,125], mounting mechanisms [126], as well as measurement instruments [127] could help immensely in improving the economic viability of agrivoltaic systems.

4.4.5. Legal and Regulatory Frameworks

Legal, regulatory, and governance frameworks surrounding agrivoltaic decommissioning and residual land use should be critically examined. Researchers should engage with policymakers to explore whether current solar end of life regulations adequately account for the dual-use value of agrivoltaics, and to propose updated permitting or tax schemes that support passive climate adaptation infrastructure.

4.4.6. Climate Change Scenarios

Lastly, as climate change advances, predictive models that integrate agronomic and energy outcomes under future temperature, precipitation, and solar radiation regimes should be developed. These models would allow researchers and decision-makers to evaluate the potential adaptive value of unpowered agrivoltaics under mid- and high-emission climate scenarios for 2050 and 2100.

5. Conclusions

This study demonstrates that agrivoltaic systems, even after the cessation of power generation, can continue to deliver meaningful agronomic and economic value through passive shading. At the system level, lifespan analysis revealed that while inverters typically fail within 5–20 years, PV modules remain electrically viable for at least 25–35 years and mechanically viable for up to ~1000 years, and racking and foundations can endure for 40 to more than 100 years depending on material design. This finding underscores the long-term shading potential of existing infrastructure, suggesting that decommissioned or unpowered arrays should not be dismissed as obsolete but rather reconsidered as durable agronomic assets. Analogous to shelterbelts or shade nets, these systems can continue to support soil conservation, reduce crop heat stress, and sustain livestock grazing even beyond their energy-producing lifespans.

Experimental findings showed that unpowered arrays provided stronger soil cooling effects than powered ones, with reductions of up to 11.9 °C beneath 44% transparent modules compared to unshaded controls. Soil moisture retention also improved, with volumetric water content increasing from 0.231 m³/m³ in control plots to as high as 0.245 m³/m³ under 8% modules. Air temperature reductions of up to 6.8 °C further demonstrate the potential of these systems to mitigate microclimate extremes. Such cooling effects have important implications for soil health, nutrient cycling, and crop productivity, reinforcing the role of unpowered agrivoltaics as a passive climate adaptation tool.

Yield modeling further indicated that these microclimate benefits can translate into substantial economic gains, with positive implications for farm resilience under increasingly volatile climate and market conditions. For example, unpowered agrivoltaic shading was associated with additional farm income of USD $3.9 billion for potatoes in Germany, and USD $32 billion for tomatoes in the U.S., with global potential exceeding USD $580 billion annually across major crops. These results illustrate the significant economic co-benefits of retaining PV infrastructure as a form of passive agricultural shading.

Policy frameworks must adapt to this dual-use reality. Current solar decommissioning regulations, which focus narrowly on electrical end-of-life management, overlook the enduring utility of PV infrastructure in agriculture. Integrating agronomic co-benefits into decommissioning policy could help reduce regulatory and economic barriers, while supporting long-term farm productivity, biodiversity, and climate resilience.

Nevertheless, future research should therefore expand multi-season trials, rigorously compare powered versus unpowered configurations under identical conditions, evaluate mechanical integrity over decades, and integrate life-cycle economics with climate adaptation modeling. In sum, the evidence presented here highlights the need to rethink agrivoltaics not solely as an energy technology but as a multi-functional agricultural infrastructure. By recognizing the post-electricity utility of these systems, policymakers, farmers, and researchers can unlock new pathways for sustainable land use, resilient food systems, and climate adaptation.