
==== Front
Heliyon
Heliyon
Heliyon
2405-8440
Elsevier

S2405-8440(24)12089-0
10.1016/j.heliyon.2024.e36058
e36058
Research Article
Vegetable crop growth under photovoltaic (PV) modules of varying transparencies
Hickey Thomas
Uchanski Mark mark.uchanski@colostate.edu
⁎
Bousselot Jennifer
Department of Horticulture and Landscape Architecture, Colorado State University, USA
⁎ Corresponding author. 1173 Campus Delivery, Fort Collins, CO, USA. mark.uchanski@colostate.edu
09 8 2024
30 8 2024
09 8 2024
10 16 e360588 4 2024
8 8 2024
© 2024 The Authors
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
The present study summarizes two growing seasons (2020–2021) of microclimate characterization and vegetable crop growth in an agrivoltaics system in northern Colorado, USA. The replicated experiment evaluated three module transparency types (opaque silicon [0 % transparent], bifacial silicon [∼5 % transparent], and semi-transparent cadmium telluride [40 % transparent]) plus a full sun control, and four vegetable crop species (summer squash, peppers, tomatoes, and lettuce). Air temperature under the modules in July was approximately 0.5 °C cooler than in the full sun. Soil temperature (2.5 cm depth) maximum differences were more pronounced and were 5.8 °C, 9 °C, and 14.4 °C cooler under bifacial, semi-transparent, and opaque silicon, respectively. For summer squash growing directly under the solar modules, yield was significantly reduced under each of the module transparency types. However, there was no statistically significant yield reduction for peppers, tomatoes, and lettuce indicating their suitability in an agrivoltaics system. The numerical yield of most crops increased as the transparency of the solar modules increased, which could be the focus on future work.

Graphical abstract

Image 1

Keywords

Agrivoltaics
Microclimate
Agriculture
Crop
Colorado
==== Body
pmc1 Introduction

The drivers of agrivoltaics (APV) development are multifaceted, encompassing energy transition objectives, land-use efficiency, crop performance, economic benefits, and climate resilience. In the US alone it is predicted that 10 million acres of land-based photovoltaics (PV) will be installed as the country moves towards a decarbonized grid by the year 2050 [1]. Land scarcity and competition for agricultural land have motivated researchers and practitioners to explore ways to maximize land-use efficiency [2,3]. APV systems research is important to avoid conflicts between sustainable development goals.

APV allows for the productive use of land by harnessing solar energy while simultaneously supporting agricultural activities [4]. Researchers have identified key barriers to adoption from the agricultural sector that can be used to guide research objectives to optimize the mutual benefits of APV [5]. A report from the National Renewable Energy Laboratory (NREL) highlights the 5 C's of Agrivoltaic Success, bringing technical and social considerations to the forefront of APV development - Climate, Configuration, Crop, Compatibility and Collaboration - essential components of a larger framework for successful APV integration [6]. By sharing the same land, there is opportunity for farmers to generate additional income from solar power while preserving the primary agricultural function within the framework of the larger Food Energy Water Nexus (FEWN; [6]).

Shading from solar modules in APV systems reduces evaporation rates, preserving soil moisture and enhancing water-use efficiency. By mitigating water loss through evaporation, APV systems help alleviate the strain on water resources, especially in arid and semi-arid regions [7]. Many studies have found increased soil moisture and reduced soil temperature within APV systems ([8]; Uchanski et al., 2023, [9]). Other studies have indicated that the reduced evaporation in APV systems contributes to improved water-use efficiency, allowing crops to grow with reduced irrigation requirements [7,10].

In addition, the integration of solar modules/panels within agricultural landscapes creates a microclimate that shields crops from extreme weather events such as hail, wind, and excessive UV radiation. This semi-protective microclimate can mitigate crop damage by producing more stable growing conditions [11]. While some studies report profound microclimatic effects from the PV canopy in their climate, it is important to account for local climatic conditions when interpreting data [8].

Sunlight is fundamental to both agricultural and PV systems and can be the primary factor influencing plant success in APV systems. The positioning and arrangement of solar modules within these systems can significantly impact the availability and distribution of sunlight reaching the underlying crops. Several studies across the world have quantified the impacts of shade from PV on crop systems ([12,13], [14]).

APV systems implemented globally predominantly utilize conventional opaque silicon PV modules, which can significantly alter the microclimate beneath the modules, particularly under high shading ratios [12]. Semi-transparent PV (STPV) module technology has emerged as a potential solution to mitigate the negative effects of dense shade in cropping systems while maintaining a high module density. While STPV modules have not been studied at the same rate as opaque technologies, they offer several opportunities to provide optimal plant growth conditions in APV systems through tunable transparencies, spectrum splitting technologies, and building-integrated solutions [12]. With further research specific STPV technologies will likely emerge as optimal solutions for distinct crop types and climates across the globe.

1.1 Crop yield

The impact of varying shade conditions from PV modules on crop yield has been studied for more than 10 years with wide-ranging results depending on crop type, APV system configuration, location, and climate. Dozens of different crop types and cultivars have been studied in APV systems including vegetative crops and fruiting crops ([7,12,13,15,16], [17])

Field trials and experimental studies have demonstrated that different plant species exhibit varying responses to alterations in light conditions associated with different regional climates and APV configurations [18,19]. In an APV system with elevated modules in Arizona, researchers found a fruit count increase in pepper cultivars [7], while a review of several crop studies reported general trends towards yield reduction with greater shading density [13]. Solar modules in APV systems provide shading, mitigating the negative impacts of excessive heat on crops brought on by predicted climate change. Reduced heat stress can improve crop yields, as higher temperatures often hinder physiological processes and limit photosynthetic activity [7].

The present study summarizes two growing seasons (2020–2021) of microclimate characterization and vegetable crop growth in an APV system in northern Colorado, USA. Four species of vegetable crops were grown under fixed-tilt solar arrays with three module transparency types – opaque silicon, bifacial silicon, and semi-transparent cadmium telluride. This research aims to investigate and analyze the application of semi-transparent module technology in APV systems, focusing on microclimate changes and crop yield response when compared to conventional opaque PV modules and a full sun control. To our knowledge this is the first replicated research experiment that evaluates module transparency types in an irrigated vegetable field setting. This work contributes to the body of work in the expanding topic of agrivoltaics.

2 Materials & methods

2.1 Site Description

This study was carried out during the 2020 and 2021 growing seasons at the CSU Agricultural Research, Development, and Education Center, South (ARDEC S.), which is located at coordinates 40.610012, −104.993979, with an elevation of 1523 m above sea level. The ARDEC S. site in Fort Collins, CO, encompasses 3.4 ha of certified organic land specifically designated for research and demonstrations related to vegetable cropping systems. The study utilized a permanent experimental installation of 9 pole-mounted PV arrays on the certified organic land at ARDEC S. as described in the field study in Uchanski et al. [20].

The soil at ARDEC S. is classified as Nunn clay loam [21]. Prior to planting each year, soil samples were collected from a depth of 20–30 cm and analyzed at the CSU Soil, Water, and Plant Testing Laboratory to measure pH, electrical conductivity, lime, texture, organic matter, and nutrient content to determine the appropriate fertilizer rates for the growing season.

2.2 PV Description

The PV plots were arranged in a randomized complete block design (RCBD) with three different module transparency types and one full sun control, replicated three times in an open field setting (Fig. 1). The first was a thin film, semi-transparent cadmium telluride CdTe (ST-CdTe), which had a transparency of 40 % (Fig. 1). These modules were manufactured by Advanced Solar Power in Hangzhou, China, with a rated output of 57 W. The module dimensions measured 1200 mm in length, 600 mm in width, and 7 mm in thickness. The second module type was an opaque polycrystalline silicon (O–Si), model JKM325PP-72 by Jinko Solar (Shanghai, China), which had 0 % transparency (Fig. 1). These modules had a rated output of 325 W, with dimensions of 1956 mm in length, 992 mm in width, and 41 mm in thickness. The third module type was a bifacial monocrystalline silicon (BF–Si), model LR6-72BP-360M by Longi, which had a transparency of approximately 5 % and a rated output of 360W. (see Table 1)Fig. 1 Monofacial polycrystalline module (0 % transparent), Bifacial monocrystalline module (∼5 % transparent), Cadmium telluride (CdTe) thin-film module (40 % transparent).

Fig. 1

Table 1 Basic information about the PV modules and system used in this study.

Table 1PV module description	% light transmission	Commercially available in the US?	
Opaque silicon	0	Yes	
Bifacial silicon	5	Yes	
Semi-transparent cadmium telluride	40	No	

Each of the nine PV arrays were mounted on the Montana Solar Top-4 racking system (Fig. 1). The ST-CdTe arrays consisted of a combination of six landscape-oriented and four portrait-oriented modules, providing a comparable surface area to the silicon arrays (Fig. 1, lower right image). The racking system for each array was fixed to a single steel pole mount with a diameter of 152.5 mm, which was installed into a concrete pad measuring 600 mm in width and 1830 mm in depth. The array's tilt angle could be manually adjusted, ranging from a vertical position (∼0°) (Fig. 2) to a nearly horizontal position (90°), or parallel to the soil surface. The adjustable tilt allowed for farm equipment to pass close to the array for field preparation, and then can be angled flat to maximize crop protection or maximize shade for farmer welfare during the harvest season.Fig. 2 Array adjustability for field preparation.

Fig. 2

Throughout the growing season, the arrays were set at an angle of 35° facing south. At this angle, the bottom edge of the modules was positioned 1220 mm above the ground, while the back edge reached a height of 2360 mm above the ground. Each of the 12 crop subplots, including both PV arrays and control plots, spanned a width of 4.3 m, with a 4.3 m spacing between adjacent subplots. Due to the single pole mount configuration the shadow cast from the modules moved throughout the day. With this, the crops received direct sun early and late in the day, with maximum shade during the peak hours of the day and immediately under the modules. The design of the arrays aimed to replicate field conditions encountered in open field APV operations [20], however financial constraints and experimental layout created edging effects that should be noted.

2.3 Plant material

The crop species tested in 2020 included two pepper cultivars (Capsicum annuum ‘Ace F1’ and 'Early Jalapeño'), summer squash (Cucurbita pepo ‘Early Prolific Straightneck’), and lettuce (Lactuca sativa ‘Capitata’). In 2021 the peppers, lettuce, and squash cultivars remained the same to complete the two-year study. In addition, two tomato cultivars (Solanum lycopersicum ‘Tasmanian Chocolate’ and ‘Red Racer’) were grown in 2021 for one season of data.

2.4 Planting and management

There were three planted rows across the entire site - north, middle, and south (Fig. 3). Lettuce, peppers, and tomatoes were planted in two, offset sub-rows in 0.9 m beds covered with black plastic mulch in the north and south rows. Squash was exclusively planted in the middle row both years with 1.2 m spacing on center. In 2020 the crops in the north and south rows were planted with a consistent planting plan in each of the 12 replications, while in 2021 the crops in the north and south rows were planted with randomized location within each replication. The north and south rows each had one line of drip irrigation buried under black plastic mulch, while the middle row was supplied with two lines of surface drip irrigation (i.e. no black plastic mulch present). Irrigation was supplied at 15 Lpm/30.5 m for 3-h intervals three days a week across all plots. There were two fertigation events in each of the growing seasons using Drammatic ONE Fish Emulsion (4-4-0.5, N–P–K) at a rate of 150 ppm.Fig. 3 ARDEC South APV research site planted with open bed in the middle row, and plastic mulch in the north and south rows.

Fig. 3

2.5 Data collection

Weather data from 2020 to 2021 was recorded at the CSU ARDEC Main CoAgMet station. The maximum (max), minimum (min) and average (avg) air temperature and relative humidity data for the months of June, July, August and September are shown below (Fig. 4).Fig. 4 ARDEC weather data for northern Colorado in the 2020 and 2021 growing seasons.

Fig. 4

Crop parameters measured included total yield in weight per harvest per plant for all crops plus individual fruit count per squash plant. Environmental parameters measured included air temperature at 30 cm above the soil surface modules, and soil temperature 2.5 cm under soil surface (Fig. 5). Soil and air temperatures were measured at 30-min intervals over the course of the entire growing season using HOBO H21-USB micro station data loggers (Onset Computer Corporation; Bourne, MA, USA).Fig. 5 Air and soil temperature data collection points for the APV array located in northern Colorado, USA.

Fig. 5

Crop yield was routinely evaluated as crops reached harvestability, which was dependent on the crop type and growing habits. Squash yield was collected once a week consistently for 5 weeks after the initial harvest (8/14/2020, 7/22/2021). Lettuce yield was measured only once per year by destructively harvesting the crops 6 weeks after planting (6/30/2020, June 7, 2021). Pepper yield was collected on a rolling basis every two weeks after the initial harvest (9/16/2020, January 9, 2021) until the end of the growing season and a killing frost. Tomatoes were harvested every two weeks after the initial harvest (8/17/2021). Total crop weight (g) was collected at each harvest as well as the plant count, and average weight per plant was calculated. The average weight per plant accounted for plant mortality throughout the growing season.

We report the crop yield from summer squash, lettuce, jalapeño peppers, and bell peppers as a combined two-year study (2020–2021), while two tomato cultivars were added in the second year (2021). All yield findings are presented as the result of the four treatments replicated three times in space: full sun (control), O–Si, BF–Si, and ST-CdTe.

2.6 Data analysis

R-studio (2002, Boston, MA) was used to analyze the results of our data collection using two separate linear mixed models - one for squash, and one for all other crops in the north and south rows to test for differences among treatments. Pairwise comparisons of mean yield/plant using the Kenward-Roger method and Tukey adjustment were used to compare a family of four estimates for significance at a 95 % confidence interval.

2.7 Light analysis

Seasonal light conditions were modeled using Ladybug Tool's Honeybee plug in for Rhinoceros 3D software (Fig. 6)(www.ladybug.tools, Robert McNeel & Associates) to analyze the differences in total average solar radiation that reached the crop canopy, or ground level under each module type. The average irradiance over the course of the growing season – May –Fig. 6 Visualization of the modeled direct and diffuse irradiance in this system using Ladybug Tools for Rhinoceros 3D.

A) Aerial Perspective of the 3D model space with row outlines for reference on the right side. B) Plan view output visualizing differences in average ground level irradiance from May 1 – October 31 between the 3 module types (O–Si [0 % transparent], BF–Si [5 % transparent], and ST-CdTe [40 % transparent]). The right side includes row planting outlines for spatial reference. C) A comparison of ground level irradiance under the O–Si array in June, September, and December.

Fig. 6

October – in Fort Collins, CO is shown in Fig. 6a and b. Fig. 6c shows a comparison of irradiance under the O–Si arrays during the three months that align with the summer solstice, fall equinox, and winter solstice to highlight how the shadow moved throughout the solar year.

Throughout this study, the south row of crops never received shade from the PV (Fig. 3, Fig. 6b). Crops grown in the north and south rows were combined in the analysis, as this is how crops would be planted in a ground-mounted, fixed-south array configuration. The authors also note that all treatments likely experienced edge effect due to the nature of the single pole mounted structures. All treatments received direct light in the morning and the evening, but remained shaded during the hours surrounding solar noon every day. While the array configurations were fixed south, the diurnal light pattern closely resembled that of a single axis tracking system. As a result, our results reported here describe this specific system's unique characteristics.

2.8 Energy yield analysis

Yearly PV power generation data was gathered and summarized remotely using SolarEdge inverters (SolarEdge Technologies, Inc.; Herzliya, Israel) and web-based software (https://monitoring.solaredge.com/) for opaque silicon and bifacial silicon (Fig. 7). Power generation data was not available for the semi-transparent cadmium telluride modules since they were not UL-listed for grid connection at the time of this publication.Fig. 7 Yearly PV power generation (kWh) from a replicated agrivoltaics array in northern Colorado, USA. Opaque silicon modules are dark blue (BF–Si 1,2,3) and bifacial silicon are light blue (O–Si 1,2,3).

Fig. 7

3 Results & discussion

3.1 Temperature

3.1.1 Air temperature

In Fort Collins, Colorado, July is the hottest month on average [22] and therefore most likely to show a microclimate impact under the PV modules. In the present study, the measured maximum, minimum, and average air temperature (Table 2), and soil temperature (Table 3) for the month of July in 2021 are shown here. On average, during the month of July, the air temperature at 30 cm above the soil was 0.4 °C cooler under the ST-CdTe treatment compared to the full sun treatment (control), and 0.6 °C and 0.5 °C cooler under O–Si and BF–Si, respectively (Fig. 8). Temperature is an important factor in crop plant function, health, and yield. Sustained extreme temperatures can cause changes in physical and biochemical plant properties that can impact all stages of growth, from germination to photosynthesis and reproduction [23].Table 2 Air temperature under APV arrays in July of 2021 in northern Colorado.

Table 2July Air Temperature	Full Sun Temp, °C	ST-CdTe Temp, °C	O–Si Temp, °C	BF–Si Temp, °C	
July Max	40.7	39.0	38.1	38.4	
July Min	10.2	10.2	10.3	10.5	
July Mean	23.2	22.8	22.6	22.7	
ST-CdTe = semi-transparent cadmium telluride, O–Si = opaque silicon, BF–Si = bifacial silicon.

Table 3 Soil temperature under APV arrays in July in northern Colorado.ST-CdTe = semi-transparent cadmium telluride, O–Si = opaque silicon, BF–Si = bifacial silicon

Table 3July Soil Temperature	Full Sun Temp, °C	ST-CdTe Temp, °C	O–Si Temp, °C	BF–Si Temp, °C	
July Max	40.7	31.7	26.3	34.9	
July Min	16.7	15.8	15.7	15.0	
July Mean	25.4	22.1	21.1	23.1	

Fig. 8 Soil temperature under APV arrays in the first week of July 2021 ST-CdTe = semi-transparent cadmium telluride, O–Si = opaque silicon, BF–Si = bifacial silicon.

Fig. 8

3.1.2 Soil temperature

The average soil temperature at a 2.5 cm depth was 3.3 °C cooler under the ST-CdTe treatment compared to the full sun treatment, and 4.3 °C and 2.3 °C cooler under O–Si and BF–Si respectively (Table 3). Soil temperature maximum differences at the same depth were more pronounced and were 5.8 °C, 9 °C, and 14.4 °C cooler under BF–Si, ST-CdTe, and O–Si, respectively.

These results are comparable to Marrou et al. [16], where they found soil temperature was cooler, specifically 0.5 °C–2.3 °C, in shaded areas when compared to non-shaded soil portions in the APV system. Similar to our findings, no significant differences in air temperature were reported [16]. In the same study in France, Marrou et al. (2013b) found more profound differences between shade and non-shaded soil temperature compared to air temperatures.

A study in Italy also found an average decrease of 1 °C under APV systems [15]. The authors attributed the reduction in solar radiation to the decrease in mean soil temperature in addition to evapotranspiration and water balance.

Others have suggested that variation in air temperature in APV systems may be dependent on the shading pattern from the system configuration [24]. APV system configuration including inter-row spacing, tracking or fixed, elevation above the ground, module transparency, and other factors can all impact the shading density and pattern throughout the day which can impact other microclimatic conditions like temperature and soil moisture [10]. Research conducted in Arizona's arid climate found that crops sustained less heat and drought stress under the shade of their elevated APV structure [7]. The microclimatic effects from these systems can play a significant role in the reduction of evapotranspiration, increase photosynthetic efficiency, and therefore reduced irrigation needs in dry regions.

3.2 Crop yield

3.2.1 Summer squash from the center row

Summer squash was grown in the center row, directly under the arrays (Fig. 9). In the combined analysis of the 2020 and 2021 growing seasons, the squash yielded 5.1 kg per plant in full sun control treatment, 3.2 kg in the BF–Si treatment, 3.2 kg in the O–Si treatment, and 4.1 kg in the ST- CdTe treatment (Fig. 10). For summer squash, the location of the row down the center point of the experimental array resulted in significantly lower yields under the PV regardless of the module transparency type.Fig. 9 Summer squash under ST-CdTe array.

Fig. 9

Fig. 10 Average squash yield (g) per plant. Means were significantly different in all treatments when compared to the control. BF–Si (p = .002), O–Si (p = .003), ST-CdTe (p = .045).

Fig. 10

There have not yet been any published studies reporting squash yield in APV systems. However, Rogers [25] completed a consumer study on several crops that were grown in an APV system, including winter squash. In this study they found that squash samples grown in the shade of PV modules were slightly preferred by consumers, which could be attributed to perceivable differences in color and sugar content. Future research could include both squash yield and quality.

3.3 Crops in the north and south rows

3.3.1 Bell pepper & jalapeño pepper

We found numerically equal or higher average yield in both pepper varieties in the ST-CdTe treatments. However, no statistically significant differences between treatments were found, indicating that all panel treatments resulted in similar yield compared to full sun (i.e. no yield loss under the modules). Over two years, the jalapeño peppers yielded 155 g per plant in full sun control treatment, 161 g in the BF–Si treatment, 155 g in the O–Si treatment, and 162 g in the ST- CdTe treatment (Fig. 10). Over two growing seasons, the bell pepper variety yielded 295 g per plant in full sun control treatment, 294 g in the BF–Si treatment, 278 g in the O–Si treatment, and 346 g in the ST- CdTe treatment (Fig. 11). When grown in the north and south rows of this experiment, bell and jalapeño peppers did not experience a significant yield reduction under the three module types of differing transparencies. In the semi-arid climate of the Negev Desert, sweet peppers cultivated under moderate shade conditions (12–26 % reduction of full sunlight) exhibited enhanced yields and increased plant heights [26].Fig. 11 Average pepper yield under different module transparencies. Means were not significantly different at p < .05.

Fig. 11

3.3.2 Lettuce

We found that lettuce yield was numerically higher under PV modules, but did not differ significantly between any of the treatments and full sun control. Over both growing seasons, the lettuce fresh weight equaled 105 g per head in full sun control treatment, 126 g in the BF–Si treatment, 111 g in the O–Si treatment, and 129 g in the ST- CdTe treatment (Fig. 12).Fig. 12 Average lettuce yield. Means were not significantly different at p < .05.

Fig. 12

In one of the first studies on lettuce growth in APV systems Marrou et al. [27] reported mixed results depending on the season of planting and the density of PV modules. They found a significant reduction in lettuce yield one season in full density shade, but that was not the case in the spring season. The lettuce in the present study was transplanted in the early summer season (i.e. June) and did not show a yield reduction under the PV modules.

3.3.3 Red Racer & Tasmanian Chocolate tomato

In this portion of the study, we found no significant differences in tomato yield between the four treatments. We found the highest numerical yield mean in the St-CdTe treatments, similar to the other crops grown in the north and south rows in this two-year study. The cultivar ‘Tasmanian Chocolate’ yielded an average of 926 g per plant in full sun control treatment, 1060 g in the BF–Si treatment, 1069 g in the O–Si treatment, and 1278 g in the ST- CdTe treatment (Fig. 12). The Red Racer tomato cultivar yielded an average of 867 g per plant in full sun control treatment, 733 g in the BF–Si treatment, 903 g in the O–Si treatment, and 962 g in the ST- CdTe treatment (Fig. 13). When grown in the north and south rows, these two cultivars of tomatoes did not experience a significant yield reduction when grown under PV modules.Fig. 13 Average tomato yield under PV modules of varying transparency. Means were not significantly different at p < .05.

Fig. 13

While we report fruit yield in fresh weight, not fruit count, Barron-Gafford et al. [7] reported a doubling of cherry tomato fruit count in the shade treatment of their experiment in Arizona. These results demonstrate the opportunity that APV systems can provide in semi-arid regions of the world when the crop type and the PV configuration are considered in combination with the climate. Future research could include measures of fruit count and total yield.

Al-Agele et al. [28] studied tomatoes growing in three different locations within a fixed south APV system in the state of Oregon. They found crop yield decreased as shade increased. They note that fixed south systems have a large amount of heterogeneity in yield, which is likely due to the distinct light patterns in these systems [28]. Several other studies involving tomatoes in APV systems have been conducted in controlled environments [19,29] with mixed results depending on percentage of light reduction.

There have been other studies on tomato yield under partial shade conditions in semi-arid climates, however, these studies use shade netting instead of PV modules as the source of the shade. The results from these studies have shown that fruit yield exhibits an increase under moderate shading conditions (35 % reduction of full sunlight) in semi-arid regions characterized by high light intensities [[30], [31], [32]].

4 Conclusions

In this study we report temperature and crop yield response in APV under three types of PV arrays in comparison to full sun. During the hottest month of the year, maximum air temperatures were reduced under all PV types, and an even greater reduction in soil temperatures was recorded under the modules. The alteration of the growing environment through reduction of temperature encourages further investigation into plant response to the unique microclimate in APV systems. The specific yield results varied depending on the crop type, and location within the APV system, which can be attributed to the varying light conditions under the different treatments and row placement. The shade of the modules likely had the greatest impact on the squash yield as it was planted in the middle row directly under the APV arrays, while all other crops grown in the north and south rows did not experience as much impact from shade and therefore, we did not find any significant reduction in yield.

The results of this study demonstrate the potential to optimize solar racking and semi-transparent module technologies to match the ideal conditions for a specific specialty crop system. The optimization of the APV array with semi-transparent PV modules could increase agricultural production while maintaining the added protection of an energized canopy in traditional APV systems. More research is needed to better understand the economic tradeoffs between increased module transparency compared to vegetable crop production, while also considering the increased energy yield from module bufaciality. Further research should explore modeling, different module transparencies and configurations effect on crop yield, soil moisture, and evapotranspiration rates in semi-arid climates.

CRediT authorship contribution statement

Thomas Hickey: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing – original draft, Writing – review & editing. Mark Uchanski: Conceptualization, Methodology, Resources, Supervision, Writing – review & editing. Jennifer Bousselot: Formal analysis, Investigation, Resources, Supervision, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
==== Refs
References

1 U.S. Department of Energy The Solar Futures Study 2021 National Renewable Energy Lab. (NREL) Golden, CO (United States) NREL/FS-6A20-80826
2 Dinesh H. Pearce J.M. The potential of agrivoltaic systems Renew. Sustain. Energy Rev. 54 2016 299 308 10.1016/j.rser.2015.10.024
3 Trommsdorff M. Dhal I.S. Özdemir Ö.E. Ketzer D. Weinberger N. Rösch C. Agrivoltaics: solar power generation and food production Solar Energy Advancements in Agriculture and Food Production Systems 2022 Academic Press 159 210 10.1016/B978-0-323-89866-9.00012-2
4 Dupraz C. Marrou H. Talbot G. Dufour L. Nogier A. Ferard Y. Combining solar photovoltaic panels and food crops for optimising land use: towards new agrivoltaic schemes Renew. Energy 36 10 2011 2725 2732 10.1016/j.renene.2011.03.005
5 Pascaris A.S. Schelly C. Pearce J.M. A first investigation of agriculture sector perspectives on the opportunities and barriers for agrivoltaics Agronomy 10 12 2020 1885 10.3390/agronomy10121885
6 Macknick J. Hartmann H. Barron-Gafford G. Beatty B. Burton R. Seok-Choi C. Davis M. Davis R. Figueroa J. Garrett A. Hain L. Walston L. The 5 Cs of Agrivoltaic Success Factors in the United States: Lessons from the InSPIRE Research Study 2022 National Renewable Energy Lab. (NREL) Golden, CO (United States) 10.2172/1882930 No. NREL/TP-6A20-83566)
7 Barron-Gafford G.A. Pavao-Zuckerman M.A. Minor R.L. Sutter L.F. Barnett-Moreno I. Blackett D.T. Thompson M. Dimond K. Gerlak A.K. Nabhan G.P. Macknick J.E. Agrivoltaics provide mutual benefits across the food–energy–water nexus in drylands Nat. Sustain. 2 9 2019 848 855 10.1038/s41893-019-0364-5
8 Marrou H. Dufour L. Wery J. How does a shelter of solar panels influence water flows in a soil–crop system? Eur. J. Agron. 50 2013 38 51 10.1016/j.eja.2013.05.004
9 Williams H.J. Hashad K. Wang H. Zhang K.M. The potential for agrivoltaics to enhance solar farm cooling Appl. Energy 332 2023 120478 10.1016/j.apenergy.2022.120478
10 Adeh E. Selker J.S. Higgins C.W. Remarkable agrivoltaic influence on soil moisture, micrometeorology and water-use efficiency PLoS One 13 11 2018 e0203256 10.1371/journal.pone.0203256
11 Bousselot J. Slabe T. Klett J. Koski R. Photovoltaic array influences the growth of green roof plants Journal of Living Architecture 4 3 2017 9 18 10.46534/jliv.2017.04.03.009
12 Gorjian S. Bousi E. Özdemir Ö.E. Trommsdorff M. Kumar N.M. Anand A. Kant K. Chopra S.S. Progress and challenges of crop production and electricity generation in agrivoltaic systems using semi-transparent photovoltaic technology Renew. Sustain. Energy Rev. 158 2022 112126 10.1016/j.rser.2022.112126
13 Laub M. Pataczek L. Feuerbacher A. Zikeli S. Högy P. Contrasting yield responses at varying levels of shade suggest different suitability of crops for dual land-use systems A meta-analysis. agriRxiv 2021 2022 20210479141 10.1007/s13593-022-00783-7
14 Trommsdorff M. Kang J. Reise C. Schindele S. Bopp G. Ehmann A. Weselek A. Högy P. Obergfell T. Combining food and energy production: design of an agrivoltaic system applied in arable and vegetable farming in Germany Renew. Sustain. Energy Rev. 140 2021 110694 10.1016/j.rser.2020.110694
15 Amaducci S. Yin X. Colauzzi M. Agrivoltaic systems to optimise land use for electric energy production Applied energy 220 2018 545 561 10.1016/j.apenergy.2018.03.081
16 Marrou H. Guilioni L. Dufour L. Dupraz C. Wery J. Microclimate under agrivoltaic systems: is crop growth rate affected in the partial shade of solar panels? Agric. For. Meteorol. 177 2013 117 132 10.1016/j.agrformet.2013.04.012
17 Tajima Makoto Iida Tetsunari Evolution of agrivoltaic farms in Japan AIP Conference Proceedings 2361 2021 030002 10.1063/5.0054674
18 Aroca-Delgado R. Pérez-Alonso J. Callejón-Ferre Á.J. Velázquez-Martí B. Compatibility between crops and solar panels: an overview from shading systems Sustainability 10 3 2018 743 10.3390/su10030743
19 Touil S. Richa A. Fizir M. Bingwa B. Shading effect of photovoltaic panels on horticulture crops production: a mini review Rev. Environ. Sci. Biotechnol. 20 2 2021 281 296 10.1007/s11157-021-09572-2
20 Uchanski M. Hickey T. Bousselot J. Barth K. Characterization of agrivoltaic crop environment conditions using opaque and thin-film semi-transparent modules Energies 16 2023 3012 10.3390/en16073012
21 Natural Resources Conservation Service; U.S Department of agriculture. Soil 2018: 2018 web soil survey Available online: https://websoilsurvey.sc.egov.usda.gov/ 24 February 2018
22 NOAA’s National Weather Service Climate 2023 https://www.weather.gov/wrh/climate?wfo=bou
23 Wahid A. Gelani S. Ashraf M. Foolad M.R. Heat tolerance in plants: an overview Environ. Exp. Bot. 61 3 2007 199 223
24 Weselek A. Ehmann A. Zikeli S. Lewandowski I. Schindele S. Högy P. Agrophotovoltaic systems: applications, challenges, and opportunities. A review Agron. Sustain. Dev. 39 2019 1 20 10.1007/s13593-019-0581-3 30881486
25 Rogers M. Consumer Study of Agrivoltaics Food Products Including Tomato, Basil, Potato, Bean, and Squash (Doctoral Dissertation, the University of Arizona) 2022 Available from wilProQuest Dissertations & Theses Global 2672269026
26 Rylski I. Spigelman M. Use of shading to control the time of harvest of red-ripe pepper fruits during the winter season in a high-radiation desert climate Scientia horticulturae 29 1–2 1986 37 45 10.1016/0304-4238(86)90029-4
27 Marrou H. Wéry J. Dufour L. Dupraz C. Productivity and radiation use efficiency of lettuces grown in the partial shade of photovoltaic panels Eur. J. Agron. 44 2013 54 66 10.1016/j.eja.2012.08.003
28 Al-agele H.A. Proctor K. Murthy G. Higgins C. A case study of tomato (Solanum lycopersicon var. Legend) production and water productivity in agrivoltaic systems Sustainability 13 5 2021 2850
29 Waller R. Kacira M. Magadley E. Teitel M. Yehia I. Semi-transparent organic photovoltaics applied as greenhouse shade for spring and summer tomato production in arid climate Agronomy 11 6 2021 1152 10.3390/agronomy11061152
30 Kittas C. Katsoulas N. Rigakis V. Bartzanas T. Kitta E. Effects on microclimate, crop production and quality of a tomato crop grown under shade nets J. Hortic. Sci. Biotechnol. 87 1 2012 7 12 10.1080/14620316.2012.11512822
31 Nangare D.D. Singh J. Meena V.S. Bhushan B. Bhatnagar P.R. Effect of green shade nets on yield and quality of tomato (Lycopersicon esculentum Mill) in semi-arid region of Punjab https://krishi.icar.gov.in/jspui/handle/123456789/14937 2015
32 Masabni J. Sun Y. Niu G. Del Valle P. Shade effect on growth and productivity of tomato and chili pepper HortTechnology 26 3 2016 344 350 10.21273/HORTTECH.26.3.344
