Greenhouse crop production in northern countries often relies heavily on supplemental lighting for year-round yield and product quality. Among the different spectra used in supplemental lighting, red is often considered the most efficient, but plants do not develop normally when grown solely under monochromatic red light (“red light syndrome”). Addition of blue light has been shown to aid normal development, and typical lighting spectra in greenhouse production include a mixture of red and blue light. However, it is unclear whether sunlight, as part of the light available to plants in the greenhouse, may be sufficient as a source of blue light. In a greenhouse high-wire tomato (Solanum lycopersicum), we varied the percentage of blue supplemental light (in a red background) as 0, 6, 12, and 24%, while keeping total photosynthetically active radiation constant. Light was supplied as a mixture of overhead (99 μmol m-2 s-1) and intracanopy (48 μmol m-2 s-1) LEDs, together with sunlight. Averaged over the whole experiment (111 days), sunlight comprised 58% of total light incident onto the crop. Total biomass, yield and number of fruits increased with the addition of blue light to an optimum, suggesting that both low (0%) and high (24%) blue light intensities were suboptimal for growth. Stem and internode lengths, as well as leaf area, decreased with increases in blue light percentage. While photosynthetic capacity increased linearly with increases in blue light percentage, photosynthesis in the low blue light treatment (0%) was not low enough to suggest the occurrence of the red light syndrome. Decreased biomass at low (0%) blue light was likely caused by decreased photosynthetic light use efficiency. Conversely, decreased biomass at high (24%) blue light was likely caused by reductions in canopy light interception. We conclude that while it is not strictly necessary to add blue light to greenhouse supplemental red light to obtain a functional crop, adding some (6–12%) blue light is advantageous for growth and yield while adding 24% blue light is suboptimal for growth.

No abstract

In the past decade, the potential of positioning LED lamps in between the canopy (intra-canopy) to enhance crop growth and yield has been explored in greenhouse cultivation. Changes in spatial heterogeneity of light absorption that come with the introduction of intra-canopy lighting have not been thoroughly explored. We calibrated and validated an existing functional structural plant model (FSPM), which combines plant morphology with a ray tracing model to estimate light absorption at leaflet level. This FSPM was used to visualize the light environment in a tomato crop illuminated with intra-canopy lighting, top lighting or a combination of both. Model validation of light absorption of individual leaves showed a good fit (R2 = 0.93) between measured and modelled light absorption of the canopy. Canopy light distribution was then quantified and visualized in three voxel directions by means of average absorbed photosynthetic photon flux density (PPFD) and coefficient of variation (CV) within that voxel. Simulations showed that the variation coefficient within horizontal direction was higher for intra-canopy lighting than top lighting (CV=48% versus CV= 43%), while the combination of intra-canopy lighting and top lighting yielded the lowest CV (37%). Combined intra-canopy and top lighting (50/50%) had in all directions a more uniform light absorption than intra-canopy or top lighting alone. The variation was minimal when the ratio of PPFD from intra-canopy to top lighting was about 1, and increased when this ratio increased or decreased. Intra-canopy lighting resulted in 8% higher total light absorption than top lighting, while combining 50% intra-canopy lighting with 50% top lighting, increased light absorption by 4%. Variation in light distribution was further reduced when the intra-canopy LEDs were distributed over strings at four instead of two heights. When positioning LED lamps to illuminate a canopy both total light absorption and light distribution have to be considered.

No abstract