Leading researchers and practitioners of LED-based horticultural lighting gathered Oct. 9, 2018 in Portland, OR for the third LEDs Magazine Horticultural Lighting Conference. The full house had great networking opportunities, the tabletop exhibits (Fig. 1) were packed during breaks and the evening reception, and the speakers delivered a wealth of knowledge relative to the science behind solid-state lighting (SSL) for horticulture, as well as the business case for artificial lighting and especially LED-based lighting. Indeed, there were compelling talks on plants’ response to spectrum, new metrics, reading plant feedback, and the SSL business case; we will summarize some of those presentations here.
The keynote speaker at the 2018 conference was Erik Runkle (Fig. 2), professor and extension specialist in the department of horticulture at Michigan State University. Runkle was an early pioneer in using LEDs for horticultural lighting, and he has perhaps the most advanced SSL-based horticultural lighting laboratory in the world. We covered details of the seven-channel LED system Runkle uses in the Controlled Environment Lighting Laboratory (CELL) in a prior article and only provide brief details here. It also features on this issue cover. The system (Fig. 3) includes two phosphor-converted white channels, four monochromatic color channels, and an ultraviolet (UV) channel.
Excluding or supplying wavelengths?
Runkle began his talk explaining that his Masters and PhD research was focused on light and how to regulate light in a greenhouse. “In the late 1990s or so, there were LEDs around, but no one imagined they would be coming to horticulture as soon as they have. So a lot of my research focused on how light regulated flowering and how different wavebands, when we excluded that, regulated growth and development. We’re almost doing the opposite now. Instead of excluding different wavebands, we are now delivering intentionally different wavebands. And the fun part is that we are learning what different wavebands do, but we can’t always predict then how they react with each other.”
Runkle also lamented that real examination of intensity was still in the very early stages, and research thus far has focused just on light, whereas it may be important to ultimately study how light interacts with other environmental parameters. He added, “That’s going to play a bigger role in our future research.”
Early in his talk, Runkle introduced a relatively simple graphic (Fig. 4) that summarized the three ways in which light and the application of light can impact plants. He described the impact as representing three dimensions: They are light quantity or intensity, light quality or spectral power distribution (SPD), and duration of light application per day that’s often described as the photoperiod of the lighting system. Primarily, those dimensions impact plant biomass, morphology, and flowering, respectively.
Plant biomass refers to the growth of the shoots or the roots, or thickness of the stem. Morphology refers to the physical architecture of the plant such as how tall is the stem and how many branches are present. Flowering is self-explanatory, although Runkle explained that goals for flowering can be decidedly different for, say, an ornamental relative to a vegetable.
FIG. 2. Erik Runkle from the Michigan State University horticultural department delivered the keynote address.
Runkle said it would be really convenient if research could treat the dimensions in silos, each relative to the primary outcome attributed above. But he noted, “The more we learn about lighting and light applications for plants, the more we realize that these factors all interact with each another.” That factor, illustrated in the graphic, complicates research and effectively means that researchers have appropriate controls to study all of the dimensions — thus the complexity of the SSL installation in Runkle’s lab.
PAR and more
Next. Runkle discussed how plants react to light in general, starting with the inevitable discussion of horticultural-specific metrics, and ultimately building to the complex set of graphs shown in Fig. 5. He said, “It’s natural that we apply our own bias to the perception of light to plants.” As we have written many times previously, that’s simply not an appropriate concept. The aforementioned graph still shows the human visual response to light labeled as luminous efficiency (LE).
The graph makes it clear that Runkle’s work has exceeded the bounds of the so-called photosynthetically active radiation (PAR) region that stretches between the 400–700-nm wavelengths. Runkle has done research focused both on the UV region below the PAR region and the far-red region above the PAR region. He said research into far red is in the relatively early stages, but such wavelengths do impact photosynthesis.
FIG. 3. Runkle’s Controlled Environment Lighting Laboratory (CELL) features a seven-channel SSL system developed for the research work by Osram. [Photo credit: Michigan State University CELL, photo by Qingwu (William) Meng.]
Ultimately, what the graph shows is that there are multiple photoreceptors in plants that must be targeted with horticultural lighting; this launched that discussion focused on phytochrome absorption. There are two different types of phytochromes — one that absorbs in the red spectrum at the top end of the PAR band (PR) and one that absorbs in the far-red region above the PAR band (PFR). The absorption peaks are at 660 and 735 nm, respectively, and Runkle said these receptors have been documented in most plants. The far-red photoreceptors impact plant reactions such as shade avoidance responses like elongation of stems or expansion of leaves.
Another important receptor or pigment found in all plants is the cryptochrome. Absorption peaks for the cryptochrome receptors vary by cultivar, according to Runkle, but the graph he showed located cryptochrome reception generally in the blue region and also in the UV band. He said the 450-nm spectrum generally delivers good cryptochrome response. Ultimately, Runkle said there are far more photoreceptors, but he focused his presentation on the blue, red, and far-red bands as shown in Fig. 5.
Runkle said the prevalence of blue, red, and far-red LEDs has made it convenient to study plant responses to energy in those bands. But he also dismissed the idea that green light is not efficient in terms of causing chlorophyll absorption. He showed a graph of photon reception across bands from 300–800 nm, and only slightly more light is reflected or transmitted in the green region relative to the blue or red regions. Much of Runkle’s research is focused on what positive impacts can be achieved with far-red spectrum and what are the impacts of including green spectrum in the mix.
Applications and timing
Runkle also stressed that research needs to consider whether we are measuring the effect of photons on an instantaneous basis, or whether we are considering long-term applications and how plants react to such longer exposure. Runkle said he is generally focused more on the long-term reaction.
He said the general applications of horticultural lighting fall into one of three categories:
- Photoperiodic lighting in greenhouses or outdoors to impact a specific action such as flowering
- Supplemental lighting in greenhouses to increase growth and yield
- SSL indoors such as in vertical farms
Shown in a simple experiment with photoperiodic lighting for flowering, the Campanula ornamental remained in a vegetative stage with a photoperiod of 14 or fewer hours of light. But a photoperiod of 15 hours or more yielded a reproductive plant that was flowering. Runkle described that particular impact as very dramatic but said many times the expected impact will be far more subtle.
FIG. 4. Three dimensions of light produce different impacts on plants, but researcher are also learning that there is interaction among the dimensions.
Originally, much of the flowering research was done with incandescent lighting because such sources produce both red and far-red spectrum, and researchers validated that adding far-red spectrum improved flowering. Now Runkle and others can use LEDs to more granularly test red to far-red ratios. Indeed, he stated that LED lighting is not effective at accelerating flowering unless far-red spectrum is present. In the case of research with snapdragons, he said a far-red to red ratio of 0.6–0.7 provides the best results. Understand that ratio indicates a greater amount of far-red energy than red energy is needed to impact flowering.
Supplemental greenhouse lighting
Moving to supplemental greenhouse lighting, Runkle said you primarily still see high-pressure sodium (HPS) lighting today because the fixtures are very affordable, although he said LED adoption has increased quite a bit in the last few years. The goals in such applications are increasing yields for something like tomatoes or developing thicker stems, and an overall higher-quality plant with ornamentals.
The most important metric in such supplemental lighting is the total amount of light provided to the plant quantified as the daily light integral (DLI). In a simple example, he showed ornamental seedlings grown under 6, 10, and 16 DLI measured in mol/m2×d. That range might correspond to a cloudy fall day in the Pacific Northwest or to a sunny day in a warmer climate. The longer DLI results in thicker stems and better rooting — and the better likelihood that the plant can survive transportation and transplantation.
Runkle said the positive impact of supplemental lighting is inarguable. But the value proposition is not. A grower must consider the cost of supplemental lighting and whether a faster growth cycle and superior plant can justify the cost. And there is no simple answer. As for whether SSL is viable, he said, “It’s a question I get quite a bit. The answer is it depends. That’s not a cop-out; it’s just that you have to look at your specific growing conditions, the crops that you are producing, what time of year, how expensive is your electricity, etc.”
Does spectrum matter?
Of course, a key question is did the spectrum matter — the answer to which might favor LED lighting over legacy sources. Runkle said that in tests with petunias under HPS and LED sources, including different mixes of red, green, and blue spectrum with the LEDs, intensity mattered far more than spectrum when the light provided was in the range of a photosynthetic photon flux density (PPFD) of 90 µmol/m2×s. But he said spectrum appears to matter more when growers use far lower intensity.
FIG. 5. There are far more photoreceptors in plants than originally thought. The graph shows the human vision response labeled as luminous efficiency (LE) relative to the response of phytochrome red (PR) and far-red (PFR) pigments, and the cryptochrome (CRY) pigment, all relative to readily available monochromatic LED spectrum.
Lastly, Runkle turned his attention to sole-source lighting including vertical farms. And the research spanned older work as well as research in the relatively new CELL facility mentioned previously. The researchers had tested geraniums with different mixes of blue, red, and far-red LEDs, and with variations in intensity of each of the three. The results are far too complex to detail here. But in terms of plant health and quality, the far-red spectrum generally improved the results and the blue spectrum was important as well. In subsequent tests, the addition of green spectrum was evaluated in lettuce. Both green and far-red spectrum were shown to increase leaf expansion. Runkle concluded that while optimum spectrum is “highly situational,” it can have “pronounced effects” in sole-source lighting applications. That said, there is spectral interaction that needs much more study and the interaction with intensity needs more study also.
Metrics, standards, and incentives
Moving on, there were a number of compelling talks at the conference that focused on metrics, standards, and emerging regulatory guidance that will impact whether horticultural luminaires qualify for market-transformation incentives such as utility rebates. Axel Pearson provided a comprehensive update on the DesignLights Consortium’s (DLC) latest work centered around horticultural lighting that will be a significant gate to incentive programs. We covered that work in a prior article and won’t detail it here.
Standards development also remains important and is the basis for programs such as the DLC. We published another recent article detailing the work of the American Society of Agricultural and Biological Engineers (ASABE) on the new ANSI/ASABE S642 standard defining methods for testing LED products relative to plant growth.
The business case
Now, let’s turn to the business case. Several speakers at the conference presented case studies with payback analysis. In many cases, LEDs still lose out to HPS and other legacy sources because the legacy sources cost far less and, in some cases, produce greater intensity. The Lighting Research Center (LRC) published research last year making just such a point regarding horticultural lighting. But LEDs are on a constant improvement path.
Doug Oppedal, principal at Evergreen Consulting Group, provided one such case study relative to cannabis — a cultivar where HPS remains the choice of many growers. He specifically studied a flowering room at a commercial growing operation, and flowering is the stage where the greatest intensity is needed. The study considered a room with 218 1000W HPS lights compared to the same number of 640W LED lights that cost $1200 each. In Oregon, such an SSL investment would garner just over $100,000 in incentives. With the incentives accounted for, the payback is 2.4 years and from that point forward the investment would deliver $64,128 in additional profits each year.
One of the most anticipated presentations of the day came at the close of the conference from Philip Smallwood, director of research at Strategies Unlimited and co-chair of the conference (Fig. 6). Indeed, one attendee noted on the conference evaluation that was turned in at the end of the day that he “Loved him some Philly Smallwood.” At the conference, Smallwood presented his most recent market research on the horticultural lighting application. (You’ll find the full market update and forecast report available for purchase on the Strategies Unlimited website at http://bit.ly/2CLCnIa.)
FIG. 6. Philip Smallwood said LEDs will penetrate 30–40% of the greenhouse lighting market by 2022 but would surpass legacy sources in cannabis growing applications by 2020.
Smallwood first sought to establish a sense of reality around the application, noting that many people are predicting the market for LED-based horticultural lighting could grow to $8 billion in five years. He said, “It is an exciting application, and it is going to grow, but there is a lot of detail that needs to go behind these numbers.” He said you need to understand how different applications will have different levels of acceptance for SSL, where are LEDs being used around the world, and that there are geopolitical issues with cultivars such as cannabis. Indeed, just geography determines how much supplemental lighting might be required.
He also said we have to remember that “this is a business.” All decisions about using LEDs in horticulture need to be based on a cost analysis. And Smallwood and his team factor payback into their market projections. He explained that there are undeniable benefits to LED lighting. Tunable spectrum is one. And that doesn’t necessarily mean tunable during operation but instead in the luminaire manufacturing process when the LED mix can be customized. Moreover, the long life and the ability to place the sources near the plant are very big advantages. You can’t place legacy sources in such proximity because of the heat generated. Still, he said, “I have no doubt that LEDs are going to take over this market. It’s just a matter of how quickly they are going to do it.”
Available market by footprint
The Strategies Unlimited research is based on an analysis of the available market, which is ultimately the footprint of potential horticultural lighting applications. For example, Smallwood said global greenhouse space totals 55 billion ft2 today. But less than 1% of that total uses supplemental lighting. Regions near the equator may not need supplemental lighting and poorer economic regions simply can’t afford it.
Certainly, regions such as North America and Western Europe can afford greenhouses and supplemental lighting. Smallwood said the total installed base of greenhouses in the US is about 1.5 billion ft2, and that 10–15% of that space uses supplemental lighting including ornamental and vegetable growing operations. Western Europe has a larger base with about 20 billion ft2 and about 10% of that using supplemental lighting.
Today, the penetration of LEDs in ornamental and vegetable operations remains very low. But Smallwood said he sees that as a positive because it represents opportunity for LEDs. The greenhouse market is currently a $4 billion market globally, growing to $8 billion by 2022. LED-based lighting will represent the majority of that growth, climbing to 30–40% of the installed base by 2022.
Turning to vertical farms, the market will be almost solely served by LEDs. That application will serve big population bases with limited arable land. But costs are high. Smallwood said more than 56% of vertical farm costs come from labor. Most existing vertical farms are not profitable; the ones that are have taken on average seven years to become profitable, according to Smallwood. He said, “It’s a slower moving market, there’s a lot of interest in it, but it’s not there yet.” He said the lighting market for vertical farms will total more than $350 million by 2022.
The wild card, of course, is cannabis. “I do see that state by state, the states are accepting cannabis, not only for medical, but they also start accepting it for recreational use,” noted Smallwood. The same applies for Canada. He said licenses for production facilities in Canada have doubled in the past year and 30 million legal consumers live in Canada today, equating to 2.4 million pounds of consumption annually. Following the numbers, Canada will need 6.5 million ft2 to serve the demand, and the country further hopes to export the product.
Turning to payback, Smallwood said the spot index for cannabis was $1200 per pound. He continued with the assumption that you could increase yield by 5% via LED spectrum optimized for the cultivar. That level of efficiency equates to payback inside one year. The key is proper implementation. Growers can’t simply replace HPS fixtures with LEDs but rather must adapt their operations for the characteristics of SSL.
“For cannabis, LED lighting is going to penetrate on a quicker scale,” said Smallwood. The growth will come both outdoors in greenhouses with the illegality lifted and indoors using modified vertical farm techniques. The propagation stage of growth will move first, followed by vegetative, and flowering will be slower to adopt LEDs because of the need for higher-intensity light. LEDs will surpass legacy sources in the cannabis application in 2020, with the market for LED lighting in Canada and the US to be over $400 million.
So horticultural lighting remains a challenge with huge potential for LED manufacturers and the companies that master the art of effective luminaire design. But the science is constantly changing with new research and an expansion of the wavebands being studied. We certainly look forward to following the technology and market.