Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Advertisement
Scientific Reports volume 15, Article number: 43717 (2025) Cite this article
Agri-Photovoltaic (APV) systems combine electricity generation and agricultural production on the same land. The physiological impacts of the shading imposed on crops cultivated under such systems are not fully understood. This study evaluated the impact of APV shading on tomato physiology and productivity through two field experiments conducted in 2022 and 2023 at Bar Ilan University, Israel. Processing tomato plants (Heinz 1648, Heinz 4107) were grown in seven north–south oriented rows (T1–T7) situated between two photovoltaic (PV) panel arrays. The arrays consisted of 24 east–west-facing, single axis sun-tracking PV modules installed 1.7 m above the ground. Fruits were harvested 95–98 days after planting. Results demonstrated a strong positive correlation between total photosynthetic irradiance and tomato productivity. Shading imposed by the PV modules reduced chlorophyll content, total biomass, fruit yield, and fruit quality. Plants in the central row T4, receiving highest light availability within the APV system (1.96% less light compared to adjacent open fields), served as a control. Plants grown directly under the PV modules in rows T1 and T7 experienced the strongest shading and recorded the greatest yield losses (42% and 57%, respectively). Plants in rows with moderate shading T2 and T6 had yield losses of 13% and 20%, respectively, while plants in rows receiving near-full sunlight (T3 and T5) exhibited minimal losses (0% and 6%, respectively). Total fruit yield loss across all seven rows was 19.4% compared to conventional cultivation. Reducing the number of rows between PV modules from seven to six or five decreased yield losses to 13.0% and 7.2%, respectively, and improved the land equivalent ratio (LER). The PV systems generated ~ 29,000 kWh/1000 m² per growing season and ~ 70,000 kWh/1000 m² during the offseason. Notably, the annual net profit from tomato production under APV was 9.54 times higher than conventional agriculture.
Climate change, combined with rising global population, has intensified the need for sustainable energy alternatives to replace fossil fuels and ensure food security^1,2,3,4. Agri-Photovoltaic (APV) crop cultivation has emerged as a promising solution to meet the growing global demand for renewable energy while addressing increasing competition for land resources^{5,6,7,8,9,10,11}. APV offers a unique opportunity to integrate crop production with energy generation on the same land, a concept first proposed in 1982¹². Beyond its agricultural benefits, APV contributes to climate change mitigation by reducing greenhouse gas emissions associated with fossil-fuel-dependent farming, aligning with global initiatives such as the European Union’s goal of achieving net-zero emissions by 2050^1,2,3.
APV systems, particularly those equipped with sun-tracking photovoltaic (PV) panels, allow dynamic shading adjustments throughout the day, optimizing both agricultural and energy outputs^{4,13,14,15,16}. However, despite their potential, further research is needed to fully understand the effects of APV shading on various crops, especially those less tolerant to shade. The impact of APV shading on crop morphology, photosynthesis, and water productivity varies based on plant species, environmental conditions, and module design¹⁷.
Tomato (Solanum lycopersicum), one of the world’s most commercially significant crops, serves as an ideal model for studying the effects of APV systems on food production. Some studies suggest that shading from PV modules can reduce heat stress and water loss, improving water-use efficiency, while others indicate that excessive shading negatively affects fruit yield and quality¹⁸. Understanding how different crops respond to agrivoltaics conditions is crucial for optimizing land use and ensuring food security in a changing climate.
This study aims to assess how APV systems support the dual goals of renewable energy generation and sustainable agriculture. To achieve this, two field experiments (spring 2022 and spring 2023) were conducted at the APV Farm at Bar-Ilan University, Israel. The research examined the effects of PV modules on tomato physiology and productivity, measured plant irradiance levels, and monitored electricity generation.
Figure 1 provides a schematic overview of the APV farm at Bar-Ilan University, Israel (32.072N, 34.847E). Bar Ilan University regulations enabled the allocation of this field to conduct agricultural research to the benefit of the public. The 714 m² field was covered with white plastic mulch (0.1 mm thick, Polytiv, Einat, Israel) to prevent weed growth (Fig. 2A). A total of 500 polystyrene growth containers (1.2 m l × 0.5 m w × 0.22 m d) (Polyvid, Haifa, Israel) were arranged in 20 rows, with 25 containers per row, placed end-to-end. Of these 20 rows, two sets of 7 rows (T1–T7) were used for experimentation. Each set of 7 rows was positioned between each pair of APV modules. Rows were spaced 0.8 m apart (Fig. 2B, C). The outer rows on either end of the experimental field were utilized as guard rows.
Two-dimensional script of the BIU APV farm.
Tomato production in the APV facility at Bar Ilan University, Ramat Gan, Israel. (A) The soil between AP modules is covered with plastic sheets to avoid weeds. (B) polystyrene growth containers filled with soil mixture arranged in 7 rows between PV arrays. (C) Tomato plants grow in polystyrene containers 30 days after planting. (D) appearance of mature tomato plants at 90 days after pl anting. (E) drown view over the BIU-APV farm before harvest.
The containers were filled with “Green Compost” (Nativ-Recycling Ltd, Israel), a mix of 25% peat, 25% volcanic stones (0.4 cm size), and 50% compost. The compost contained 1.37 kg/m³ nitrogen, 0.68 kg/m³ phosphorus, and 0.27 kg/m³ potassium. Each container held 110 L of soil mixture, and five tomato plants were grown per container. An aerial view of the field at early growing stage is shown in Fig. 2C, while Fig. 2D Fig. 2E depicts the plants at harvest.. Tomato (Lycopersicon esculentum L) seedings are commercially available to the public. They were purchased from Hishtil Nurseries Ltd, Nehalim, Israel and planted in the APV Farm. Table 1 provides details on the tomato cultivars, planting dates, and harvest times.
The characteristics and configuration of the PV modules are detailed in Table 2. Three north–south-oriented PV arrays were installed in the field, each 29 m long, carrying 25 east–west-facing PV modules. The modules were inclined at 205° eastward and mounted 1.7 m above the ground, equipped with single-axis sun trackers. Four additional module arrays were installed along the fences surrounding the field (Fig. 1).
As can be judged from Fig. 1, GCR is 26%, namely, 26% of the ground area is shaded by the PV panels when panels facing the sky at full horizontal position.
Continuous irrigation and fertigation were controlled via a computerized system. Each row was equipped with two 16 mm-diameter drip irrigation pipes (Netafim Ltd, Israel) with 30 cm-spaced drippers, each delivering 1.6 L/h. The irrigation system operated continuously all through the season , four to six times a day, for 10 or 20 min, depending on the growth stage. To prevent soil salinization, irrigation was applied with a 15% leaching fraction, and excess water drained from the southern end of the field.
Fertigation commenced 30 days after planting. The following doses were applied per 1000 m² per season: 35 kg nitrogen, 6 kg phosphorus (equivalent to 13.8 kg P₂O₅), 50 kg potassium (equivalent to 60 kg K₂O), along with 0.5% magnesium, 2% calcium, 300 mg/kg iron, 150 mg/kg manganese, 75 mg/kg zinc, 11 mg/kg copper, and 8 mg/kg molybdenum.
Weather sensors were installed in each row at 0.8 m above ground to log air temperature, relative humidity, net radiation (PYR-Decagon Devices Inc., Pullman, WA, USA), and photosynthetically active radiation (PAR) (MQ-306/6 quantum, Apogee Instruments, Inc., Logan, UT, USA). Data were recorded every 10 min using Apogee AT100 data loggers and stored in the cloud.
In addition, environmental monitoring was conducted using two weather stations: one located 200 m East of the APV field at 5 m above ground (BIU Meteorological Service, Israel), and the other was positioned in the center of the APV field, 5 m above the PV modules (Weather station model GW 1103, Ecowitt, Wanchai, Hong Kong). The weather station data included temperature, RH, solar radiation, UV radiation, pressure, wind, wind direction and rainfall.. The data sets that were collected during the season from the two stations were subjected to statistical differences. The comparison revealed that the weather conditions in the two locations were similar, indicating the PV panels had no significant impact on the weather conditions that occurred at 5 m above the APV field.
Electricity generation was monitored using seven meters—one per array in the field and four additional meters for the modules mounted on the fences. The specifications of the fence-mounted modules are provided in Table 2.
LER was calculated based on Mead and Willey¹⁹ as follows:
where:
({text{LER}} > {1}) indicates that combined crop and energy production is more productive than separate production.
({text{LER}} < {1}) indicates that intercropping is less productive than monoculture.
This calculation considered row 4 as a true Full Light control, despite it receiving 1.96 % lower irradiation than true Full Light plots adjacent to the APV field.
In each row (T1–T7), ten random containers (a total of 50 plants per row) were selected for measurements. Three terminal leaflets per plant were measured along the longitudinal axis using a ruler, and stem length was recorded using a tape measure.
Chlorophyll content was assessed in the third leaf below the apex, one leaf per plant, 50 plants per row using a SPAD 502 Plus Chlorophyll Meter (Spectrum Technologies, Aurora, IL, USA) at 11:00 AM.
Five random containers per row (total of 25 plants) were selected for biomass assessment. After removing fruits, plants were uprooted, cut at the stem base, and weighed for foliage biomass. Roots were washed, blotted dry, and weighed separately.
Twenty-five plants from five randomly selected containers per row were sampled. Viable flowers (with yellow turgid petals and green sepals) were counted 32 days after transplanting.
Fruit number and weight were recorded in 25 plants per row from five randomly selected containers. Fruits were sorted into ripe and unripe categories before weighing.
Twelve ripe fruits per row were randomly selected. Each fruit was quartered, and juice samples were extracted from the core using a pipette to measure total soluble sugars (TSS). Fresh and dry weights were recorded after drying the samples in a ventilated oven at 60 °C until constant weight was achieved (three days).
Blossom end rot incidence was assessed in five random containers per row (25 plants total), with healthy and affected fruits counted per plant.
In ten random containers per row (50 plants total), leaf senescence was evaluated based on the proportion of healthy green leaves versus yellowing leaves.
Ten random containers per row (50 plants total) were examined for powdery mildew (Erysiphe neolycopersici) infection, with the percentage of infected leaves recorded.
All data were analyzed using JMP-Pro-18 statistical software (https://community.jmp.com/t5/Learn-JMP-Events/New-in-JMP-18-and-JMP-Pro-18/ev-p/810062). To control the overall error rate, analysis of variance (ANOVA) was performed, followed by paired Tukey HSD tests for mean separation (p ≤ 0.05).
Sensors installed at 0.8 m above ground in the middle of row 4 (T4) provided data on the minimum and maximum daily temperatures and relative humidity during the 2022 and 2023 growing seasons (Fig. 3). Environmental data from the six sensors in the other rows were compared with those from row 4 (Fig. 4), revealing meaningful differences in microclimate.
Meteorological conditions prevailing in the APV facility at BIU Farm during spring 2022 and spring 2023, as measured in the middle row between modules at 0.8m above ground.
The microclimate prevailing between tomato plants growing between PV modules. Values represent the differential (Δ) mean temperature and mean RH between the mean temperature and mean RH as measured at 0.8 m above ground in row 4 and mean temperature and mean RH as measured at 0.8 m above ground in rows 1, 2, 3, 5, 7. Note that the mean daily temperature (unlike the mean night temperature) is lower as rows became closer to the modules (A and B) as against the daily mean RH (unlike the night RH) which became higher as plants approached the modules (C and D).
The mean daily temperature decreased gradually as the sensor’s location approached the PV modules from both sides, likely due to shading. In both years, temperatures in rows 1 and 7 (under the panels) were approximately 2.5–3.0 °C lower than in the middle of row 4. Interestingly, the mean night temperature in rows 1 and 7 was ~ 0.5 °C higher than in row 4, suggesting that the PV modules reflected heat back to the ground at night (Fig. 4A, B). A reverse trend was observed for relative humidity (RH); during the day, RH increased as sensors approached the PV modules due to shading, while nighttime changes were minor (Fig. 4C, D).
Irradiation measurements that were taken during spring 2022 and spring 2023 confirmed that irradiation reaching row 4 (T4) was lower by only 1.96% compared to the irradiation reaching the ground in adjacent two field plots, 100 m east bound or west bound. Nevertheless, and despite not being a true full-light (FL) control, row 4 served us as full light control for comparison purposes with the other six rows. Plants grown in rows 1 and 7 under the eastern and western PV modules experienced severe shading. Hourly cumulative irradiation measurements that were taken on June 8, 2023, confirmed that these rows received the least sunlight (Fig. 5A). Cumulative daily sunlight irradiation throughout the 2023 season (95 days) is shown in Fig. 5B. Maximum irradiation within the APV structure occurred in the central control row (T4) , reaching 1.100 × 10⁶ µmoles/m²/sec/season as against 1.122 × 10⁶ µmoles/m²/sec/season in the External Control plots. The relative gradual reduction in irradiation levels in rows 1–7, compared to row 4, were 92.7%, 19.3%, 11.9%, 0.0%, 8.6%, 16.5%, and 91.2%, respectively. The relative gradual shading reduction in irradiation levels in rows 1–7, compared to External Control, were 92.9%, 20.8%, 10.9%, 1.96%, 10.2%, 24.9%, and 91.3%, respectively.
Photosynthetic irradiation (400–700 nm) reaching tomato plants growing in rows 1–7 between two PV modules. Rows 1 and 7 are located underneath the east and the west module, respectively. Row 4 is equally distant from both modules. (A) irradiation at various hours of the day of June 8, 2023. (B) Total irradiation reaching the plants during 95 days of growth in spring 2023. The extent of shading during the season in rows 1,2,3,4,5,6, and 7 as calculated relative to row 4 from data in (B) was 92.7, 19.3, 11.9, 0, 8.6, 16.5 and 91.2%, respectively. The total seasonal irradiation in row 4 was 1.96% lower compared to the irradiation in two External Controls located 100m away from each side of the APV field. Dashed gray columns in (A) and (B) represent the daily and seasonal cumulative irradiations in External Controls.
Leaf traits and fruit yield showed a decreasing trend with reduced solar radiation, especially under severe shading (T1, T7).
In both years, leaflet size was affected by row position within the APV structure (Fig. 6A). Plants in row 4, which received the highest light availability within the APV system (considered by us as receiving “full” sunlight), developed the smallest leaves while plants in the most shaded rows (1 and 7) developed the largest leaves. The average leaflet size in row 4 ranged between 8.7–9.0 cm. It increased insignificantly (9.0–9.7 cm) in rows 3 and 5. As plants approached the modules in rows 2 and 6, leaflet size increased significantly to 9.7–10.1 cm. The leaflet size reached its maximum size in rows 1 and 7 ranging between 10.0–11.1 cm, 22–31% larger than in row 4.
The distance of the tomato plants from the PV modules affected their morphology. (A) and (B) mean leaflet length and stem length in plants grown in rows 1–7. Note the gradual increase in leaflet and stem size as plants approached the modules (plants in row 1 and row 7 are located beneath the eastern and western modules, respectively). Different letters on bars (capitals for 2022 and small for 2023) indicate significant differences between means (Tucky HDS test, at α = 0.05).
A similar effect of shading was recorded with the stems. Stems were significantly longer (80–90 cm) in the most shaded rows 1 and 7 as compared to the stems in row 4 (55–57 cm) (Fig. 6B). The average stem length in row 2 (61–73 cm) and 6 (63–78 cm) was not significantly different from row 4. The results indicate a shade-induced stem elongation of up to 57% under the heaviest shade.
Shading had a strong impact on chlorophyll content (Fig. 7). Chlorophyll levels were lowest in row 4 and highest in rows 1 and 7. In 2022, SPAD increased gradually and significantly as plants approached the PV modules in both directions. Thus, SPAD was 34% larger in row 1 than in row 4, reflecting an adaptive response to reduced light availability. A similar trend was recorded in 2023, except that SPAD in rows 3 and 5 did not differ significantly from row 4.
Chlorophyll content (SPAD) in leaves taken from tomato plants 50 days after planting. Note the gradual increase in chlorophyl content as plants approach the PV modules. Different letters on bars (capitals for 2022 and small for 2023) indicate significant differences between means (Tucky HDS test, at α = 0.05).
Similar results for leaf size and chlorophyll content were reported for Vitis vinifera. Shade leaves which grow in lower parts of the plant or affected by nearby plants’ shading are receiving less light and are typically larger, thinner, and darker green compared to sun leaves. They contain more chlorophyll to maximize light absorption in low-light conditions. In contrast, sun leaves are smaller, thicker, and lighter green, as they are adapted to handle higher light intensities and prevent water loss^20,21,22. Due to these facts the mesophyll cell size varies differentially and different responses in carbon assimilation per photosynthesizing cell volume occur²¹.
Total fresh foliage weight did not significantly differ across all rows (Fig. 8A). In both years, no statistical differences in foliage weight were seen between row 4 (“full light”) and rows 1 or 7 (“full shade”), indicating no effect of severe shading on total foliage weight. In contrast, root biomass was significantly reduced in rows 1 and 7 (Fig. 8B) compared to row 4, suggesting a differential response of the foliage vs the root to severe shading.
The fresh weight of the foliage (A) and the root system (B) of tomato plants grown between PV modules at the end of the growing season. Different letters on bars (capitals for 2022 and small for 2023) indicate significant differences between means (Tucky HDS test, at α = 0.05).
Flower production in both years was significantly reduced in only rows 1 and 7 while remained largely unaffected in the other rows (Fig. 9), suggesting moderate effect of partial shading on flowering. On average, plants in row 4 produced 16–17 viable flowers per plant, while those in rows 1 and 7 produced only 7–8 viable flowers. The reduction in flower production across other rows ranged between only 5–10%.
The number of viable flowers per plants at 32 days after planting. Note that the number of flowers declined significantly only in T1 and T7 in plants grown under the PV modules. Different letters on bars (capitals for 2022 and small for 2023) indicate significant differences between means (Tucky HDS test, at α = 0.05).
In 2022, plants in row 4 produced the highest number of ripe fruits (~ 35 per plant). Fruit count declined significantly as the proximity of the plants to the PV modules increased (Fig. 10A). In 2023, plants in rows 3, 4 and 5 produced a similarly high number of ripe fruits, with fruit count declining significantly in the other rows (Fig. 10A). The significantly lowest number of fruits was counted in rows 1 and 7 with an average (both seasons) of 14 fruits/plant, which infers 60% reduction relative to row 4. A similar trend was observed for fruit weight per plant (Fig. 10B). Plants in row 4 yielded the highest fruit weight (~ 3.5 kg per plant), which declined significantly and progressively in adjacent rows. The yields in rows 3 and 5 were similar to row 4 while rows 2 and 6 showed a significant reduction in yield (~ 2.6 kg/plant) compared with row 4. The significantly lowest yield was recorded in rows 1 and 7 (1.4–2.2 kg/plant) which accounts for 37.1–60.0% reduction in fruit weight relative to row 4.
Yield parameters of tomato plants grown between PV modules. (A) The number of ripe fruits per plant at the end of the season. (B) The weight of ripe fruits per plant at the end of the season. Note that the yield gradually declined as plants approached the PV modules. Different letters on bars (capitals for 2022 and small for 2023) indicate significant differences between means (Tucky HDS test, at α = 0.05).
Average yield in row 4 in 2022 and 2023 accounted for 10.5 ton/1000 m², similar to the yield obtained by farmers in Israel (Oren Barnea, personal communication; https://www.gov.il/he/pages/moag-calculation-719) and Italy²³.
The number of ripe fruits at harvest was significantly affected by shading conditions (Fig. 11). The average number of ripe fruits per plant in 2022 in rows 1, 2, 3, 4 5, 6 and 7 was 19, 27, 32, 35, 32, 27 and 19, respectively. In 2023- 10, 29, 35, 35, 35, 29 and 10, respectively. Fruit count in rows 1 and 7 was reduced significantly by 45.7% in 2022 and by 71.4% in 2023, relative to row 4, but not that in rows 2–6 (Fig. 11A).
The number of ripe and unripe fruits per plant (A) and the proportion of ripe fruits per plant (B) at the end of the season as affected by the distance of the tomato plants from the PV modules. Different letters on bars indicate significant differences between means (Tucky HDS test, at α = 0.05). Note that the number of ripe fruits was more affected than the number of unripe fruits.
The proportion of immature green fruits (unripe fruits) showed an opposite trend to that of ripe fruits. It was significantly higher in rows 1 and 7 than in the other rows (Fig. 11A). While the percentage of mature (red) fruits exceeded 90% in rows 2–6, it was reduced by nearly 50% in rows 1 and 7 (Fig. 11B).
Tomato Fruit in row 4 contained ~ 5% of dry matter. Rows 2,3,5 and 6 contained ~ 5% of dry matter indicated no significant differences between them and row 4 . In the shaded rows 1 and 7, dry matter content was reduced by ~ 50% and reached an average of 2.7% dry matter (Fig. 12A). Interestingly, unlike the dry matter concentration, the total soluble sugars levels (~ 5%) in both years remained largely unaffected by row position with a minor exception in row 1 in 2023. (Fig. 12B). Overall, it appears that TSS levels in ripe fruits are not affected by shading and plants are able to accumulate sugars in the fruits despite the reduced radiation levels.
Percent dry matter (A) and total soluble sugars content (B) in ripe tomato fruits harvested from plants grown under PV cultivation. Note the lower percentage of dry matter as plants are grown in closer proximity to the PV modules. Different letters on bars (capitals for 2022 and small for 2023) indicate significant differences between means (Tucky HDS test, at α = 0.05).
Fresh weights of fruits and foliage per plant in each row are shown in Fig. 13A, and the ratio between them provided in Fig. 13B. Unlike fruit weight, foliage weight was relatively unaffected by the proximity to PV modules, except in rows 1 and 7. Average fruit yield stands at ~ 3 kg per plant in the central rows 2–6 as against ~ 1.9 kg per plant in the shaded rows 1 and 7. A stable average foliage weight of 0.9 kg per plant was recorded in all rows, regardless of the proximity to the PV modules (Fig. 13A).
Total fruit fresh weight and foliage fresh weight per plant (A) and the ratio between them (B) in tomato plants grown under APV cultivation. Note that unlike fruit weight, foliage weight is almost unaffected by the proximity of the tomato plants to the PV modules. In Fig. 12A Different letters on bars (capitals for fruit yield and lower case for foliage weight) indicate significant differences between means (Tucky HDS test, at α = 0.05). In Fig. 12B Different letters on bars (capitals for 2022 and lower case for 2023) indicate significant differences between mean ratios (Tucky HDS test, at α = 0.05).
The fruit yield/foliage weight ratio in 2022, reached an average of ~ 3.7 in rows 2–6, significantly lower compared to ~ 2.1 in rows 1 and 7. In 2023, the yield/foliage weight ratio reached an average of ~ 2.8 in rows 2–6, significantly lower compared to ~ 1.7 in rows 1 and 7. The low yield/foliage weight ratio in rows 1 and 7 compared with rows 2–6 possibly reflects the developmental priorities of the plants when grown under different light/shade regimes. Under shortage of light photosynthetic products are preferably used to build up vegetative biomass to catch up the sunlight.
Plants in rows 1 and 7 exhibited increased susceptibility to powdery mildew (Erysiphe neolycopersici) compared with call central rows 2–6 (Fig. 14A–B). The extreme difference between the shaded rows and the central rows indicates that the poor physiological condition of the shaded plants prevents the plant from expressing its plant defense mechanisms.
Biotic (A–B) and abiotic (C–F) stress in tomato plants grown for 80 days under APV cultivation. (A and B) disease severity of powdery mildew caused by the ascomycete fungus Erysiphe neolycopersici. (C and D) severity and symptoms of Black End rot syndromes. (E and F) Percent and the appearance of leaf senescence. Note that all stress symptoms occurred in only rows 1 and 7, both under the PV modules. Different letters on bars (capitals for 2022 and small for 2023) in A, C and E indicate significant differences between means (Tucky HDS test, at α = 0.05).
Black end rot susceptibility increased dramatically in rows 1 and 7 compared with central rows 2–6 (Fig. 14C–D). Black end rots (blossom end rot) is a common physiological disease caused by a calcium deficiency in the developing fruit. This deficiency often arises from inconsistent watering, overwatering or root damage, preventing the plant from properly absorbing calcium. Moreover, if the plant experiences rapid growth during fruit set development, it may not be able to absorb enough calcium quickly enough to meet the fruit’s needs^24,25. The over vegetative growth which is expressed in low yield/foliage weight (Fig. 13B) and the reduction of root mass (Fig. 8B) may testify to the high commonality of the black end rot in rows 1 and 7 compared with the central rows.
The senescence of the lower leaves of tomato plants was observed in shaded rows 1 and 7 while in the central rows 2–6 sentence was not observed at all (Fig. 14E–F). Brouwer et al., suggest that the induction of leaf senescence during shading depends on the efficiency of carbon fixation, which in turn appears to be modulated via light receptors such as phytochrome A²⁶ Causin et al., suggest that blue light which is in absence in shaded area prevents senescence by maintaining high levels of catalase activity²⁷.
Electricity output from the PV modules (central and fence-mounted) between March 2022 and September 2023 is presented in Fig. 15A. The average monthly electricity generation was 10,535 kWh, with the three central modules contributing 62.5% of total output. Peak electricity production occurred between 11 AM and 3 PM (Fig. 15B). Sun-tracking east–west-oriented modules were the most efficient, followed by south-facing modules with tracking, while horizontally fixed modules on the east fence were least efficient (Fig. 15C).
Electricity generation by PV modules at BIU APV facility. (A) monthly electricity generation during 2022 and 2023. Green bars- by three central arrays, blue bars- by fence arrays, red bars- total electricity generation. (B) Daily electricity generation of the 7 PV arrays at 21.06.2023 (The longest day of the year). (C) Relative array efficiency compared with the central arrays.
Mean ripe fruit weight per plant for each row (T1–T7) in 2022 and 2023 is presented in Fig. 16A. Relative to row 4 (representing conventional farming), yield losses in rows 1, 2, 3, 5, 6, and 7 were 58%, 14%, 0%, 0%, 7%, and 21%, respectively. With five, six, or seven rows of tomato plants cultivated between PV modules, the mean fruit yield losses were 7.9%, 13.0%, and 19.4%, respectively (Fig. 16B, Table 3).
Calculating yield loss and LER for tomato production under APV. (A) Mean fruit yield per plant. (B) % yield under 5, 6 or 7 row cultivation between modules. (C) LER under 5, 6 or 7 row cultivation between modules.
Land Equivalent Ratios (LER) for five, six, and seven rows were 1.10, 1.18, and 1.24, respectively (Fig. 16C, Table 3), calculated as follows:
A = agriculture, V = voltaic, cp– Cultivated land portion. * % Common coverage of open PV land in Israel with a tilt of 30°.
Despite a total yield loss of 19.4%, the most efficient APV configuration was seven rows of tomato plants per module pair, due to complete land utilization.
Table 4 presents a financial analysis for a 1000 m² APV field, extrapolated from the study of 713 m² field using a factor of 1.4. After accounting for all costs, including investment, mortgage, and maintenance, the total annual net income per 1000 m² for five-row and seven-row APV systems was 7111 NIS and 7223 NIS, respectively. Compared to conventional tomato farming, the net annual income was 9.54 times higher for a seven-row APV system and 9.39 times higher for a five-row system (Table 4).
Further calculation showed that an investment made to construct APV field of 1000 m², with 26% coverage with solar panels, could be paid back after 13.5 years with tomato production, or after 13.9 years without tomato production.
Agri photovoltaics (APV), also known as Agri-PV or agrivoltaics, is an emerging field that integrates solar photovoltaic (PV) energy production with agriculture on the same land. This dual-use system optimizes land-use efficiency while addressing food-energy-water challenges, making it a crucial step toward sustainable agriculture. The widespread adoption of APV requires continued research, policy support, and technological innovations.
Globally, the agriculture and food chain sectors consume approximately 30% of total energy and contribute 19–29% of annual greenhouse gas emissions³, primarily due to the reliance on traditional fuels. APV offers a sustainable solution by integrating clean energy production into agricultural systems¹³. Currently, over 15 GW of APV capacity has been installed worldwide²⁸, and its potential to support climate change mitigation and land-use efficiency has been widely recognized. Trommsdorff et al., emphasized that APV is a viable and efficient technology for addressing 21st-century challenges, including land scarcity and climate change¹². Similarly, Ansuya et al., highlighted APV’s role in enhancing energy production and system reliability, making it an economically attractive solution²⁹. A recent European Commission report¹ acknowledged APV as an innovative method for PV deployment, recommending further interdisciplinary research into the interactions between energy production, crop yield, and biodiversity.
Despite its benefits, several challenges hinder the widespread adoption of APV, particularly for small-scale farmers. The high initial investment costs and the lack of clear regulatory frameworks pose significant barriers. One key regulatory issue is the proportion of land that can be covered by PV modules without severely impacting agricultural productivity. Magarelli et al.³⁰ suggested a 30% shading threshold to prevent significant reductions in fruit yield and quality. Mohamedi et al.³¹ reported that shading levels in tomato crops of 43% (semi-transparent APV panels) and 67% (conventional APV panels) led to yield reductions of 28% and 58%, respectively. Scarano et al.³² reported lower tomato fruit numbers under PV panels but noted increased fruit size and water content, suggesting that appropriate irrigation strategies could mitigate shading effects.
Widmer et al.¹⁸ reviewed several studies on tomato-PV intercropping^{13,32,33,34,35,36,37,38,39,40}. They concluded that yield variations depended on location and PV configuration. Friman-Peretz et al.⁴¹ used flexible semi-transparent organic photo voltaic (OPV) modules as shading elements for a greenhouse tunnel with a tomato crop. They observed that despite the lower radiation in the OPV tunnel the cumulative number of tomatoes, their mass, and average tomato mass were higher than in the control tunnel, due to a much lower canopy temperature and slightly lower air temperature. Yalcin et al.⁴² compared three APV configurations and concluded that their broader adoption has the potential to play a crucial role in the sustainable integration of food and energy systems, thereby fostering both ecological sustainability and economic resilience within the agricultural sector.
Despite projections that solar energy will become Israel’s primary source of renewable energy by the 2030s (Ministry of Energy, Israel, www.energy-sea.gov.il, 2020) its widespread implementation significantly lags national targets. Shriki et al.⁴³ developed a GIS-based land suitability analysis that determines suitable areas for installing large-scale, ground-mounted solar photovoltaic farms and their corresponding technical potential. They indicated⁴⁴ that a total area of 323 km2 is considered appropriate for PV installations in Israel, with 42.6% of that area (137.6 km2) can be classified into high or very high suitability levels. The slow implementation of APV systems probably derives, among other reasons, from insufficient data on the impacts of APV shading on crop physiology.
The present study provides a detailed analysis of shading effects imposed by PV modules on tomato growth and productivity. The results demonstrate that shading intensity varies, depending on the proximity of the crops to the PV modules. Sun-tracking PV modules created different light environments. On June 8, 2023, solar irradiation reaching tomato crops growing at 0 m, 1 m, 2 m, and 3 m from the modules measured 51, 1250, 1310, and 1400 µmol/m²/s, respectively. The strongest negative effects were observed in tomato plants directly beneath the modules, while those grown at distances of 1–3 m exhibited minor or negligible shading effects. Chlorophyll content (SPAD), but not flowering, was the major physiological trait affected by shading: it significantly increased as shading increased, compensating for light harvesting.
Two experimental seasons (spring 2022 and spring 2023) confirmed that plants growing in rows 1 and 7 (under the eastern and western modules) suffered the greatest reductions in yield, while those growing in rows 2–6 showed minimal losses. The highest fruit yield per plant (~ 3.5 kg) was recorded in row 4, which received the highest light availability within the AV system . Yield losses increased progressively in rows closer to the PV modules.
Land Equivalent Ratio (LER) calculations for this study yielded a value of 1.24, indicating a 24% improvement in land-use efficiency compared to conventional farming. A coverage of up to 26% of the land area with PV modules did not significantly affect plant growth or fruit quality. However, reducing the number of cultivated rows to 6 or 5 decreased the LER to 1.18 and 1.1, respectively.
Our economic analyses revealed that electricity sales more than compensated for these losses. Farmers adopting APV can earn an additional 7223 NIS (~ $1980) per 1000 m² annually- approximately ten times higher than conventional tomato farming. Even if rows 1 and 7 were left uncultivated, a net gain of 7111 NIS would still be achieved. Additionally, our observations indicate that shade-tolerant crops such as pineapples may be successfully cultivated in rows 1 and 7, improving overall land productivity.
In future experiments we aim to compare the efficiency of the current APV facility with a new one that will include 4 m high constructions equipped with one- or two-arrays modules.
Our findings confirm that APV enables the dual use of land for crop production and electricity generation, making it a highly profitable and sustainable technology. The broader adoption of APV has the potential to significantly reduce carbon emissions while maintaining agricultural productivity, offering substantial benefits for both farmers and the environment.
All data are available upon request from the corresponding author.
Chatzipanagi, A., Taylor, N. & Jaeger-Waldau, A. Overview of the potential and challenges for agri-photovoltaics in the European Union (Publications Office of the European Union, 2023). https://doi.org/10.2760/208702.
Book  Google Scholar 
European Commission, 2018. Communication from the commission to the European parliament, the European council, the council, the European economic and social committee and the committee of the regions. Brussels, Belgium
International Energy Agency. Net zero by 2050: A roadmap for the global energy sector (2021).
Jordan, D., Barnes, T., Haegel, N. & Repins, I. Build solar-energy systems to last—Save billions. Nature 600(7888), 215–217 (2021).
Article  ADS  CAS  PubMed  Google Scholar 
Adeh, E. H., Good, S. P. & Calaf, M. Solar PV power potential is greatest over croplands. Sci. Rep. 9, 11442. https://doi.org/10.1038/s41598-019-47803-3 (2019).
Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 
Chalgynbayeva, A., Gabnai, Z., Lengyel, P., Pestisha, A. & Bai, A. Worldwide research trends in agrivoltaic systems—A bibliometric review. Energies 16, 611. https://doi.org/10.3390/en16020611 (2023).
Article  CAS  Google Scholar 
Ghosh, A. Nexus between agriculture and photovoltaics (agrivoltaics, agri photovoltaics) for sustainable development goal: A review. Sol. Energy 266, 112146. https://doi.org/10.1016/j.solener.2023.112146 (2023).
Article  Google Scholar 
Gorjian, S. et al. Progress and challenges of crop production and electricity generation in agrivoltaic systems using semi-transparent photovoltaic technology. Renew. Sustain. Energy Rev. 158, 112126. https://doi.org/10.1016/j.rser.2022.112126 (2022).
Article  CAS  Google Scholar 
Mamun, M. A., Dargusch, P., Wadley, D., Zulkarnain, N. A. & Aziz, A. A. A review of research on agrivoltaic systems. Renew. Sustain. Energy Rev. 161, 112351 (2022).
Article  Google Scholar 
Weselek, A., Ehmann, A. & Zikeli, S. Agrophotovoltaic systems: Applications, challenges, and opportunities. A review. Agron. Sustain. Dev. 39, 35. https://doi.org/10.1007/s13593-019-0581-3 (2019).
Article  Google Scholar 
Wilson, G. M. et al. The 2020 photovoltaic technologies roadmap. J. Phys. D Appl. Phys. 53(49), 493001 (2020).
Article  CAS  Google Scholar 
Trommsdorff, M. et al. Combining food and energy production: Design of an agrivoltaic system applied in arable and vegetable farming in Germany. Renew. Sustain. Energy Rev. 140, 110694 (2021).
Article  Google Scholar 
Hassanien, R. H. E., Li, M. & Yin, F. The integration of semi-transparent photovoltaics on greenhouse roof for energy and plant production. Renew. Energy 121, 377–388 (2018).
Article  CAS  Google Scholar 
Lama, R. K. & Jeong, H. Design and performance analysis of foldable solar panel for agrivoltaics system. Sensors 24, 1167. https://doi.org/10.3390/s24041167 (2024).
Article  ADS  PubMed  PubMed Central  Google Scholar 
Sidhik, S. et al. Two-dimensional perovskite templates for durable, efficient formamidinium perovskite solar cells. Science 384(6701), 1227–1235 (2024).
Article  ADS  CAS  PubMed  Google Scholar 
Stallknecht, E. J., Herrera, C. K. & Yang, C. Designing plant–transparent agrivoltaics. Sci. Rep. 13, 1903. https://doi.org/10.1038/s41598-023-28484-5 (2023).
Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 
Masabni, J., Sun, Y., Niu, G. & Del Valle, P. Shade effect on growth and productivity of tomato and chili pepper. HortTechnology 26(3), 344–350. https://doi.org/10.21273/HORTTECH.26.3.344 (2016).
Article  CAS  Google Scholar 
Widmer, J., Christ, B., Grenz, J. & Norgrove, L. Agrivoltaics, a promising new tool for electricity and food production: A systematic review. Renew. Sustain. Energy Rev. 192, 114277 (2024).
Article  Google Scholar 
Mead, R. & Willey, R. W. The concept of a ‘land equivalent ratio’ and advantages in yields from intercropping. Exp. Agric. 16(3), 217. https://doi.org/10.1017/S0014479700010978 (2008).
Article  Google Scholar 
Dörken, V. M. & Lepetit, B. Morpho-anatomical and physiological differences between sun and shade leaves in Abies alba Mill.(Pinaceae, Coniferales): A combined approach. Plant, Cell Environ. 41(7), 1683–1697 (2018).
Article  PubMed  Google Scholar 
Théroux-Rancourt, G. et al. Analyzing anatomy over three dimensions unpacks the differences in mesophyll diffusive area between sun and shade Vitis vinifera leaves. AoB Plants 15(2), plad001 (2023).
Article  PubMed  PubMed Central  Google Scholar 
Wu, Y., Gong, W. & Yang, W. Shade inhibits leaf size by controlling cell proliferation and enlargement in soybean. Sci. Rep. 7(1), 9259 (2017).
Article  ADS  PubMed  PubMed Central  Google Scholar 
Lovelli, S., Potenza, G., Castronuovo, D., Perniola, M. & Candido, V. Yield, quality and water use efficiency of processing tomatoes produced under different irrigation regimes in Mediterranean environment. Ital. J. Agron. 12(1), 795 (2017).
Article  Google Scholar 
Taylor, M. D. & Locascio, S. J. Blossom-end rot: A calcium deficiency. J. Plant Nutr. 27(1), 123–139. https://doi.org/10.1081/PLN-120027551 (2004).
Article  CAS  Google Scholar 
Topcu, Y., Nambeesan, S. U. & van der Knaap, E. Blossom-end rot: a century-old problem in tomato (Solanum lycopersicum L.) and other vegetables. Mol Hortic. 2(1), 1 (2022).
Article  CAS  PubMed  PubMed Central  Google Scholar 
Brouwer, B., Ziolkowska, A., Bagard, M., Keech, O. & Gardeström, P. The impact of light intensity on shade-induced leaf senescence. Plant, Cell Environ. 35(6), 1084–1098 (2012).
Article  PubMed  Google Scholar 
Causin, H. F., Jauregui, R. N. & Barneix, A. J. The effect of light spectral quality on leaf senescence and oxidative stress in wheat. Plant Sci. 171(1), 24–33 (2006).
Article  CAS  Google Scholar 
Klokov, A. V., Loktionov, E. Y., Loktionov, Y. V., Panchenko, V. A. & Sharaborova, E. S. A mini-review of current activities and future trends in agrivoltaics. Energies 169, 3009. https://doi.org/10.3390/en16073009 (2023).
Article  CAS  Google Scholar 
Anusuya, K., Vijayakumar, K., Martin, M., Leenus, J. & Manikandan, S. Agrophotovoltaics: Enhancing solar land use efficiency for energy-food-water nexus. Renew. Energy Focus 50, 100600 (2024).
Article  Google Scholar 
Magarelli, A., Mazzeo, A. & Ferrara, G. Fruit crop species with agrivoltaic systems: A critical review. Agronomy 14, 722. https://doi.org/10.3390/agronomy14040722 (2024).
Article  Google Scholar 
Mohammedi, S., Dragonetti, G., Admane, N. & Fouial, A. The impact of agrivoltaic systems on tomato crop: A case study in Southern Italy. Processes 11, 3370. https://doi.org/10.3390/pr11123370 (2023).
Article  CAS  Google Scholar 
Scarano, A. et al. Effects of the agrivoltaic system on crop production: The case of tomato (Solanum lycopersicum L.). Appl. Sci. 14, 3095. https://doi.org/10.3390/app14073095 (2024).
Article  CAS  Google Scholar 
Al-agele, H. A., Proctor, K., Murthy, G. & Higgins, C. A case study of tomato (Solanum lycopersicum var. Legend) production and water productivity in agrivoltaic systems. Sustainability 13(5), 2850 (2021).
Article  ADS  Google Scholar 
Aroca-Delgado, R., Pérez-Alonso, J., Callejón-Ferre, Á. -J. & Díaz-Pérez, M. Morphology, yield and quality of greenhouse tomato cultivation with flexible photovoltaic rooftop panels (Almería, Spain). Sci. Hortic. 257, 108768. https://doi.org/10.1016/j.scienta.2019.108768 (2019).
Article  Google Scholar 
Barron-Gafford, G. A., Pavao-Zuckerman, M. A. & Minor, R. L. Agrivoltaics provide mutual benefits across the food–energy–water nexus in drylands. Nat. Sustain. 2, 848–855. https://doi.org/10.1038/s41893-019-0364-5 (2019).
Article  Google Scholar 
Bulgari, R. et al. Micrometeorological environment in traditional and photovoltaic greenhouses and effects on growth and quality of tomato (Solanum lycopersicum L.). Ital. J. Agrometeorol. 20(2), 27–38 (2015).
Google Scholar 
Cossu, M. et al. Solar radiation distribution inside a greenhouse with south-oriented photovoltaic roofs and effects on crop productivity. Appl. Energy 133, 89–100 (2014).
Article  ADS  Google Scholar 
Ezzaeri, K. et al. Performance of photovoltaic Canarian greenhouse: A comparison study between summer and winter seasons. Sol. Energy 198, 275–282 (2020).
Article  ADS  Google Scholar 
López-Díaz, G., Carreño-Ortega, A., Fatnassi, H., Poncet, C. & Díaz-Pérez, M. The effect of different levels of shading in a photovoltaic greenhouse with a North-South orientation. Appl. Sci. 10(3), 882. https://doi.org/10.3390/app10030882 (2020).
Article  Google Scholar 
Ureña-Sánchez, R., Callejón-Ferre, Á. J., Pérez-Alonso, J. & Carreño-Ortega, Á. Greenhouse tomato production with electricity generation by roof-mounted flexible solar panels. Sci. Agric. 69(4), 233–239 (2012).
Article  Google Scholar 
Friman-Peretz, M. et al. Microclimate and crop performance in a tunnel greenhouse shaded by organic photovoltaic modules—comparison with conventional shaded and unshaded tunnels. Biosyst. Eng. 197, 12–31 (2020).
Article  Google Scholar 
Yalçın, Ö., Kuzyaka, D. & Özden, T. Agrivoltaic system design for sugar beets and wheat in Central Anatolia. Renew. Energy. https://doi.org/10.1016/j.renene.2025.122800 (2025).
Article  Google Scholar 
Shriki, N., Rabinovici, R., Yahav, K. & Rubin, O. D. Methodology for estimating the potential of ground-mounted solar photovoltaic as part of the national electricity grid: The case of Israel. Energy Sustain. Dev. 68, 332–349 (2022).
Article  Google Scholar 
Shriki, N., Rabinovici, R., Yahav, K. & Rubin, O. D. Prioritizing suitable locations for national-scale solar PV installations: Israel’s site suitability analysis as a case study. Renew. Energy 205, 105–124 (2023).
Article  Google Scholar 
Download references
This research was supported by a grant from Doral Energy Ltd., Israel.
Faculty of Life Sciences, Bar-Ilan University, 5290002, Ramat Gan, Israel
Yariv Ben Naim & Yigal Cohen
Doral Energy Ltd, Ramat Gan, Israel
Chanani Ladell
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
YBN—Investigation, Methodology, Software, Writing (original draft) CL—Resources YC—Project Administration, Supervision, Writing (review & editing).
Correspondence to Yigal Cohen.
The authors declare no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Reprints and permissions
Naim, Y.B., Ladell, C. & Cohen, Y. Agri-Photovoltaic technology allows dual use of land for tomato production and electricity generation. Sci Rep 15, 43717 (2025). https://doi.org/10.1038/s41598-025-27602-9
Download citation
Received: 25 April 2025
Accepted: 04 November 2025
Published: 12 December 2025
Version of record: 12 December 2025
DOI: https://doi.org/10.1038/s41598-025-27602-9
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative
Advertisement
Scientific Reports (Sci Rep)
ISSN 2045-2322 (online)
© 2025 Springer Nature Limited
Sign up for the Nature Briefing: Anthropocene newsletter — what matters in anthropocene research, free to your inbox weekly.

source