In a 2024 outdoor trial near Västerås, Sweden, broccoli under two semitransparent magenta solar-panel designs reached maturity 104 days after transplanting, compared with 79 days in the full-sun control—a 25-day difference. The Mälardalen University experiment tested agrivoltaics, the joint use of land for crops and solar power, at Kärrbo Prästgård. The study reports comparable mature crop measures, alongside differences in light efficiency and nutrient concentrations.
What the light-efficiency numbers measure
The panels were semitransparent cadmium telluride (CdTe) modules. Their radiation-use efficiency (RUE)—how efficiently plants turn received light into biomass—was 4.5 times the full-sun value under ASP-LAM2-T50 and 2.8 times under ASP-LAM2-T70. Those ratios describe efficiency, not total broccoli yield: the full-sun control had the higher absolute photosynthesis rate.
The manufacturer’s 50% and 70% transparency ratings also differed from the share of photosynthetically active radiation (PAR) measured reaching the crops. Photosynthetic photon flux density (PPFD) measures the flow of photosynthetically active light to the plants.
| Measure | ASP-LAM2-T50 | ASP-LAM2-T70 |
| Manufacturer transparency rating | 50% | 70% |
| Average PAR transmittance | 13.2% | 19.2% |
| Maturity after transplanting | 104 days | 104 days |
| RUE relative to full sun | 4.5× | 2.8× |
| Average maximum PPFD | 143 ± 2 μmol m⁻² s⁻¹ | 207 ± 4 μmol m⁻² s⁻¹ |
For comparison, average maximum PPFD in the full-sun control was 925 ± 17 μmol m⁻² s⁻¹. The panel plots therefore had higher RUE ratios even though the control received a higher maximum light level and had the higher absolute photosynthesis rate.
Nutrients and the land-use model
At the second sampling, micronutrient concentrations were lower beneath the panels, while macronutrient concentrations were comparable to the control. The study’s single-harvest land-equivalent ratio (LER), a modeled measure of combined land use, was 1.18 under assumptions that included a 40% allowance for land loss. It describes a modeled land-use outcome, not a measured 18% increase in broccoli yield.
The researchers also estimated that a 1.1 kWp system would generate 447.6 kWh over the cultivation period. These are study estimates, not commercial-farm performance figures.
What one plot per treatment can establish
The trial covered one broccoli cultivar, one site and one growing season, with one plot for each treatment. Six plants sampled per treatment and sampling round were subsamples, not independent plot-level replicates. The authors characterize the statistical comparisons as exploratory and subject to pseudoreplication, and call for replicated trials across seasons and locations. The experiment took place outdoors, so it did not test greenhouse performance.
