Cultivation

Lead: In the last post we went through the N/P/K numbers that actually hold up. But no feeding schedule works in isolation. The research of the past two years is clear: if you want to optimize fertilization, you have to think about light, water regime, climate and the root microbiome at the same time. And the order of those levers is not arbitrary. What this is about The first fertilization post focused on the direct N/P/K studies: Saloner and Bernstein, Bevan, Caplan. That was the groundwork. But growers in a CSC know from practice that the same ppm values produce very different results in two different setups. And that difference is not a measurement error, it is an interaction with other variables. We have widened our research library considerably since then: 50 papers in total, with a clear focus on studies that explicitly combine two or more variables. We analyzed six of them in full text. Four interaction findings came out of that, plus two negative studies that can save a club money. Water and nitrogen cannot be separated Tang and colleagues (2018) measured industrial hemp in whole-canopy gas exchange chambers, a method that captures the entire stand in real time rather than a single leaf. The result: drought stress and nitrogen deficiency reinforce each other disproportionately. Drought stress on its own lowers photosynthesis mainly through stomatal closure. Nitrogen deficiency on its own lowers it through reduced Rubisco and chlorophyll levels. When both arrive together, the plant has no reserves left. In practice that means: if you run deliberate drought stress in late flower (as Caplan and colleagues showed in 2019 for raising cannabinoid content), you need to be sure beforehand that nitrogen supply is not collapsing in parallel. Otherwise you are combining two stresses that can each be useful in the right phase, but that ruin the yield when combined. Sheldon and colleagues (2021) add the other side: drought stress tolerance varies clearly by genotype. What counts as intentional ripening stress in one cultivar is a yield loss in the next. Same rule here: observe first, standardize second. Phosphorus is the underrated lever Two recent studies suggest that common P practice is probably set too high. Shiponi and Bernstein (2021) tested five P levels (5 to 90 mg/L) on two cultivars in flower and found an opposing effect: more P raises flower yield, but lowers cannabinoid biosynthesis at the same time. That matters, because the optimum for yield and the optimum for active compounds do not coincide. If you push for maximum dry mass, you give up cannabinoid concentration. Hershkowitz, Westmoreland and Bugbee (2025) go a step further and test whether doubling the total nutrient solution concentration (from EC 2 to EC 4 mS/cm) buys anything. The answer is sober: no significant effect. Neither yield nor cannabinoids went up. More than that, the study found no further response above roughly 15 mg/L P when phosphorus was raised in isolation either. That is well below what many fertilizer manufacturers recommend on their data sheets. One qualifier: this applies to recirculating hydro setups. In drain-to-waste systems with heavy leaching, higher nominal doses still make sense, because a large share simply runs off unused. But if you recirculate or fertigate conservatively, there is money and environmental load to save here without losing quality. Light spectrum affects cannabinoids, but not all of them equally Brousseau and colleagues (2021) review the light spectrum and cannabinoid axis and deliver a differentiated picture. Cannabinoids are produced in the trichomes as photo-protective secondary metabolites. That is the mechanistic reason why manipulating the spectrum can do anything at all. UV-B can stimulate biosynthesis, visible LED spectra increase THC and terpene accumulation, CBD responds much more weakly. Morello and colleagues (2022) contribute a practical concept with “cannabinoid production efficiency”: what counts is not absolute cannabinoid content, but cannabinoid per mol of photons, in other words cannabinoid per unit of electricity invested. But: Rodriguez-Morrison and colleagues (2021) show the other side. In their controlled study, short-wave UV-B increased neither yield nor cannabinoid concentration. That is an important negative study, because plenty of commercial UV lamps are marketed on exactly that effect. The truth is probably somewhere in between: UV effects depend on cultivar, dose and growth phase, and blanket advertising claims do not survive a careful test. What does that mean for a club? If you are thinking about switching LED fixtures, cannabinoid efficiency per kWh is a more useful yardstick than the yield figure in the brochure. Microbiome: big promise, thin evidence Two reviews, Lyu and colleagues (2019) on PGPR (plant growth-promoting rhizobacteria) and Ahmed and Hijri (2021) on the microbiota in general, arrive at the same verdict: plausibly effective, but barely established causally for drug-type cannabis. The transferable evidence from rice, wheat and maize is impressive. The direct cannabis data are thin. Lyu et al. single out Pseudomonas and Bacillus strains, because the mechanisms (nitrogen fixation, phosphorus mobilization, phytohormones) are well understood, and because cannabis is a highly regulated crop for which few or no chemical plant protection products are approved in most jurisdictions, which makes biological disease control particularly attractive. One caveat: Lyu et al. is a perspective, not an original experiment, and one of the co-authors is CEO of a cannabis bioproduct company. The conflict of interest is properly documented. Even so, it is not enough as a basis for recommendations to clubs. Filho, Thomason and colleagues (2020) as well as Filho, Chim and colleagues (2024) provide original data on integrated nutrient and microbe systems that are far more concrete. Bottom line for practice: treat biostimulants as an experiment, document them, collect your own data. Read marketing claims critically. Fertilization × climate × pathogens: the unpopular coupling Punja and colleagues (2019) cataloged cannabis pathogens and molds across several sites in Canada over three years. That is the standard reference, but it is often overlooked in fertilization debates. The connection is banal and important: over-fertilization, nitrogen over-fertilization above all, produces soft, lush tissue with thin cell

Lead: LED manufacturers sell “optimised cannabis spectra” with red peaks, blue boosters and UV modules. The data says something else: spectrum effects depend almost entirely on how bright your room already is. Above 700 micromoles you can file the marketing away. Below that, it gets interesting. What this is about The first article in this series covered the big lighting levers: intensity, spectrum, UV, photoperiod. The second one went deep on photoperiod. Today it is the spectrum’s turn, and the finding is more nuanced than it looks. There is no such thing as “the right cannabis spectrum”. There are conditions under which spectrum has a strong effect, and conditions under which it barely matters at all. If you do not separate the two, you either buy needlessly expensive specialist fixtures or you miss a lever that actually works. The core insight rests on three studies we read in full: Magagnini et al. 2018 (Italy/Finland), Westmoreland et al. 2021 (USA, Utah State University) and Brousseau et al. 2021 (McGill, a review synthesis). The spectrum paradox in two studies Magagnini 2018: spectrum does everything The Magagnini group compared three light sources in a controlled tent study: an HPS lamp (the standard) and two LED spectra (AP673L, high in red, and NS1, high in blue and UV-A). All three ran at the same PPFD of 450 micromoles. The results: THC concentration under NS1 (the high-blue LED): 38% higher than under HPS in the first run, 26% higher in the second. CBG concentration: 100 to 200% higher under NS1. CBD concentration: 35% higher under LED. Flower yield: HPS slightly ahead (bigger, heavier), but the concentration advantage of the LEDs compensated for almost all of it. At 450 micromoles PPFD, spectrum matters a lot. If you run in that light class, switching to an NS1-style LED can pull noticeably more cannabinoids out of the same plants. Westmoreland 2021: spectrum does almost nothing Three years later, the Bugbee group at Utah State University measured the opposite. Both studies are methodologically sound. Westmoreland and colleagues held PPFD constant at 750 or 900 micromoles across three consecutive studies and varied only the blue fraction, from 4% (HPS) up to 20% (various LEDs). What they found: Yield dropped linearly as the blue fraction rose: a 12% drop going from 4% to 20% blue, consistent across all three studies. Cannabinoid concentration: no significant spectrum effect on THC or CBD. The white-plus-red LED setup produced 4.6% less yield per square metre than HPS, but 27% more yield per unit of electricity, because LED efficiency (micromoles per joule) was substantially better. The practical message is the opposite: at higher PPFD, spectrum is a secondary lever, and fixture efficiency matters far more economically. The resolution: photoreceptor saturation Both studies are valid within their own setups. The contradiction is only apparent, because they are measuring different things. Brousseau et al. (2021) supply the mechanistic explanation in their review: photoreceptors in the plant are not yet saturated at low light intensity. As long as cryptochrome (the blue light sensor) and phytochrome (the red light sensor) still have sensitivity headroom, they respond strongly to changes in spectrum. At high light intensity both are already largely saturated, so additional spectral variation changes little in the signalling cascade. The threshold sits somewhere around 600 to 700 micromoles PPFD. Below it, spectrum tweaks can shift cannabinoid concentrations. Above it, photon quantity dominates. That same threshold logic also explains why the UV-B findings reproduce so poorly in modern studies. The old Lydon 1987 study measured in a low-light setup, where UV-B worked as a strong stress signal. Under commercial high-output LEDs the plant is already photo-stressed enough, and additional UV-B stops making a measurable difference (Llewellyn 2022). What this means for three types of setup Hobby or small CSC room (300 to 500 micromoles PPFD) Here your choice of spectrum makes a substantial difference. If you are aiming for cannabinoid profiles, you should: Choose LEDs with a high blue fraction (15 to 25%) plus a red component Actively avoid HPS: the Magagnini data show 25 to 38% less THC under HPS at the same PPFD Treat UV-A as optional. It can help, but it is not a must Consider subcanopy lighting with red plus blue to bring lower flower tiers up (Hawley 2018, cited in Brousseau) Takeaway: at low installed power, the effort you put into LED spectrum pays off. Medium CSC or small facilities (500 to 700 micromoles PPFD) Border territory. Spectrum can still contribute something, but fixture efficiency becomes more important. The pragmatic recommendation: Look for LEDs with good efficiency (above 2.3 micromoles per joule) Do not over-weight spectrum. White-plus-red mixes with a moderate blue fraction (10 to 15%) are a robust choice Read the Westmoreland data before committing to any large spectrum investment Commercial production (above 700 micromoles PPFD) This class means licensed production, for example medical cultivation or markets outside Germany. Cultivation associations under the German KCanG are non-commercial by law. Here the logic shifts completely. Spectrum showed no measurable effect on yield or on cannabinoid concentration in these setups (Westmoreland 2021; Llewellyn 2022, who found no additional cannabinoid effect from supplemental UV under high-output LEDs). What counts instead: Fixture efficiency (micromoles per joule): the best white-plus-red LED setup returned 27% more yield per unit of electricity than HPS at the same PPFD. PPFD scaled linearly up to 1,800 micromoles without saturating (Rodriguez-Morrison 2021). Whether your room can use that headroom depends on climate control and CO2. A low blue fraction is better: 12% less yield at 20% blue than at 4% blue (Westmoreland). You can file the spectrum marketing promises away. That goes for “cannabis-specific spectra”, “UV boost” and the surcharges that come with them. Takeaway: in high-output setups, PPFD times efficiency dominates. Spectrum is fine-tuning, not an economic lever. What spectrum RELIABLY does (independent of PPFD) Three effects are stable regardless of light intensity, which makes them relevant in commercial setups too: Plant architecture: a low red-to-far-red ratio (typical of HPS) stretches stems

The reflex runs deep: rich, dark green plants look healthy, so you feed plenty of nitrogen. With cannabis that is an expensive misconception, because that deep green costs you exactly the compounds you grow for. “More is better” is the most common myth about feeding. It survives because extra nitrogen has a visible effect: more leaf mass, darker green, more vigorous growth. The problem is that visible lushness and the actual cannabinoid harvest are two different things. And a controlled trial on nitrogen for cannabis shows fairly clearly that the two can even work against each other. What the nitrogen trial shows Dilena et al. (2023) compared four nitrogen rates in clones of a CBD cultivar in a greenhouse: 30, 60, 210 and 500 mg/L over twelve weeks of flowering. The result is uncomfortable for the “more is better” reflex: Only leaf biomass rose significantly with more nitrogen, which is precisely the part you do not harvest. Cannabinoid concentration and cannabinoid yield per plant fell as the N supply went up. More fertilizer, fewer cannabinoids. The authors’ conclusion: high fertilizer rates are not recommended; the sensible corridor probably sits between 60 and 210 mg/L N. That is clean, controlled evidence for something experienced growers already suspect: in flower, cannabis is not a heavy feeder you can keep throwing N at. Past a certain point you are feeding leaves instead of flowers and pushing potency down. And pruning? It did nothing in this trial The study tested it in the same run: a double stem prune raised neither cannabinoid concentration nor biomass. That does not mean training is pointless in general, but in this setup it did not move yield or potency. A useful reality check against the many promises made for pruning and training techniques. One secondary finding is worth a look: plants at the sun-exposed edge of the greenhouse developed more biomass and higher cannabinoid concentrations. In this trial, light was a stronger lever than the fertilizer. That matches what we know from the research on light. Why too much nitrogen for cannabis costs you twice The potency loss is only half the bill. The other half is disease pressure. Punja et al. (2019) observed in their multi-year pathogen survey that overfed plants, in particular nitrogen-saturated ones with soft tissue, are more susceptible to fungal infection. Soft, N-driven tissue is easier prey for pathogens such as powdery mildew. So overfeeding can cost you twice over: once in cannabinoids, once in resilience. Put that next to Holweg et al. (2025), who found that too much heat in flower also lowers cannabinoids, and a pattern emerges: cannabis rewards restraint. Maximum input does not produce the optimal flower. The right amount does. What we do not know yet Dilena et al. tested one CBD cultivar in sand pots in a greenhouse: the exact optimal N corridor may differ with other cultivars, other substrates and with THC-dominant genetics. That 60 to 210 mg/L range is a well-founded starting point, not a law that holds across every cultivar. For now, the pruning result applies only to this one method, a double stem prune, in this setup. Other training methods were not tested. Take-away for your grow Resist the dark green reflex. The goal is not the deep green plant, it is the potent flower. And you are more likely to get there with moderate nitrogen (roughly 60 to 210 mg/L) than with the maximum dose. Too much N drives leaf mass, lowers the cannabinoids and makes the plant more prone to fungal problems. And before you bet on pruning and training techniques for extra yield: in the cleanest study available, double pruning delivered nothing. Put that energy into your light instead. Sources Dilena et al. (2023), Scientific Reports 13, 19533. doi:10.1038/s41598-023-46369-5 Punja et al. (2019), Front. Plant Sci. 10, 1120. doi:10.3389/fpls.2019.01120 Holweg et al. (2025), Environ. Exp. Bot. 230, 106085. doi:10.1016/j.envexpbot.2025.106085 → Basics: Cannabis fertilization and what the research shows

Lead: Equipment marketing promises spectrum miracles, UV boosters and perfect photoperiods. The research of the past five years paints a much clearer, almost unromantic picture: one factor beats nearly all the others, and it is not the one people talk about loudest. What this is about If you grow cannabis indoors, lighting eats the largest share of your running energy budget. So the temptation is strong to squeeze out a few more percent through spectrum tweaks, UV boosters or new photoperiods. We went through nine open-access papers published between 2019 and 2024, almost all of them from the big platforms Frontiers, PLOS and HortScience, and asked: what is solid evidence, what is marketing, and which levers are worth pulling for a grow in Germany? What works: PPFD is king If you boil the nine papers down to one statement, it is this one: more light (within sensible limits) means more flower. The relationship is surprisingly linear, and in the studies reviewed here no saturation was reported, not even at the highest levels tested. Rodriguez-Morrison et al. (2021) show this clearly in a controlled indoor study: flower mass, apical flower density and harvest index all rise linearly with PPFD. Worth noting: cannabinoid concentration did not change in this trial. So if you are chasing THC or CBD percentages, higher PPFD will not get you there. If you are chasing total cannabinoids per unit of area, it will, because more flower multiplies through. Llewellyn et al. (2022) go one step further: in a comparison of high versus low PPFD, total biomass came out 1.3 to 1.5x higher, and flower even 1.6x. Under strong light the plant does not just produce more material, it also shifts more of it into the flower (harvest index +7 %). Huber et al. (2021) remind us that light rarely works on its own: raising DLI combined with elevated CO2 produced 24 to 33 % more growth at comparable energy input. So if you are watching the electricity bill, think of light and CO2 together. Take-away: if you want to maximize one lever as a grower, PPFD or DLI is the safest bet, provided your climate management (temperature, VPD, CO2) keeps up. What matters less: the spectrum This is where it gets interesting, and uncomfortable for part of the LED industry. Westmoreland et al. (2021) held PPFD constant across three consecutive studies and varied only the blue fraction, from 4 % (HPS) to 20 % (various LED mixes). The result: more blue lowers yield linearly, by 12 % in total. In these trials, spectrum had no effect on THC or CBD concentration. Even more relevant in practice: the white+red LED setup produced 4.6 % less per square meter than HPS, but 27 % more yield per euro of electricity spent, because the LED efficacy (µmol per joule) was considerably better. In other words: the win was not the spectrum, it was light output per watt. Bilodeau et al. (2019) and Stamford et al. (2023) are both reviews with a grower focus, and they put this in context: spectrum strategies have their place, but mostly for plant shape and habit, not for a jump in yield or potency. Stamford explicitly models the electricity cost of different LED “recipes” and offers rules of thumb for when a given approach pays off. Take-away: spectrum optimization is a fine-tuning tool, not a yield lever. If you are spending money, look at fixture efficacy first. What is probably overrated: UV-B UV-B lamps have been sold as cannabinoid boosters for years. Llewellyn et al. (2022) tested this in the same study as the PPFD question and came back with a sobering answer. UV-A plus UV-B during the last 20 days of flowering did raise THC concentration in the sugar leaves (the small leaves around the flower) by roughly 30 %. But total THC per tissue stayed the same, and in the flower itself the study found no statistically significant increase in cannabinoids. The authors put it plainly in their conclusion: no commercially relevant benefit. That does not mean UV-B has no physiological effect. It means the effort does not clearly pay for itself economically, at least not in this configuration. What most people get wrong: 12:12 as gospel The standard practice of “flowering starts with 12 hours of light” comes from grow convention rather than from data, and in many cases it is suboptimal. Peterswald et al. (2023) ran nine photoperiod treatments across three cultivars: one CBD line (Cannatonic) and two THC lines (Northern Lights, Hindu Kush). The result: In all three cultivars, 14L:10D produced the highest flower yield. In the THC lines, however, THC concentration dropped under a static 14L:10D. In the CBD line everything lined up: higher yield AND higher CBD concentration, adding up to 50 to 100 % more total CBD. Ahrens et al. (2023) support that picture: in most cultivars flowering can also be triggered at 13 or 14 hours, and some manage 15. The yield response is often quadratic, with a definable sweet spot that sits between 12 and 13 hours depending on the cultivar. Take-away: photoperiod is cultivar-specific. Treat twelve hours as a universal law and, depending on the cultivar, you leave 30 to 50 % of yield or cannabinoid output on the table. Bonus: greenhouse growing For those of you who do not work purely indoors: Collado et al. (2024) looked at supplemental lighting in a greenhouse. The result: a clear increase in photosynthesis, water use efficiency and growth, with no saturation reached. Same story here: more light, more plant. Practical take-aways Lever 1 (safe): push PPFD or DLI up as far as your climate (CO2, temperature, VPD) can follow. It scaled linearly in most trials. Lever 2 (cultivar-dependent): test the photoperiod. 13 or 14 hours can deliver noticeably more flower or cannabinoids in some cultivars. Try it on a small batch first. Lever 3 (on the investment side): when buying fixtures, look at µmol per joule first, not at spectrum recipes. Lever 4 (stay skeptical):

Lead: Fertilizer advice for cannabis circulates in every forum. Solid advice is rare. Around a dozen reliable studies from the past six years now provide concrete optima for N, P and K, but they also show how strongly genotype, photoperiod, growing medium and water regime feed into the result. If you understand that, you save money and get more out of your plants. What this is about For a long time, cannabis fertilization was received wisdom: a bit of the Coffman and Gentner paper from 1977, plus experience from forums and substrate manufacturer data sheets whose recommendations rested on hearsay. That has changed since roughly 2019. Several research groups, above all the Volcani Institute in Israel (Bernstein lab) and the University of Guelph in Canada (Zheng lab), have run controlled experiments on medical drug-type cannabis. Agronomically, that is a far more reliable evidence base than anything publicly available before. For this article we evaluated twelve studies published between 2019 and 2025, six of them as full-text deep dives. Three core findings emerge. And just as important: three methodological limitations you should know about before you copy any recommendation straight into your setup. Cannabis fertilization by the numbers: N, P and K The most robust evidence exists for nitrogen in the vegetative phase. Saloner and Bernstein (2020) tested five N levels (30, 80, 160, 240, 320 ppm) on the THC/CBD cultivar “Annapurna” over 32 days. At 30 and 80 ppm the plants showed clear deficiency symptoms: chlorosis, stunted growth, less root mass. From 160 ppm upwards the optimum was reached. More did not add yield. It produced the first signs of inefficient uptake. For veg, 160 ppm N is therefore the point beyond which more stops paying off, a recommendation that appears robust across different cultivars and setups. For the flowering phase, Bevan, Jones and Zheng (2021) is the methodologically most interesting study. They used response surface methodology (a statistical approach that models interactions between variables) to determine the optimal combination of N, P and K in a hydro setup (deep water culture). The trial ran 100 plants of the THC cultivar “Gelato” over eight weeks of flower, in a Health Canada certified production room. The optimum came out at 194 mg/L N and 59 mg/L P. The interesting part: potassium was not the limiting factor within the range tested (60 to 340 mg/L), and the model could not identify a clear optimum for K. Translated into everyday practice: N and P are the dials you fine-tune in flower; K needs to stay in range, but it does not need obsessive optimization. With potassium, things become cultivar-dependent again. Saloner, Sacks and Bernstein (2019) tested two cultivars in parallel. “Royal Medic” (fairly compact) reaches its optimum at 175 ppm K, and pushing further degrades performance. “Desert Queen” (the larger growth type of the two) still benefits from an increase to 240 ppm. The study is one of the few that explicitly tests the genotype-by-fertilization interaction, and it gives the direct reason why a universal K recommendation for all cultivars cannot work. More fertilizer does not mean more cannabinoids One assumption from practice says that generous feeding gives you more THC. Bernstein et al. (2019) looked into it, with a surprising result. They tested the effect of increased N, P, K and humic acid supply on the cannabinoid profile of the cultivar “NB100” across the full vegetative and flowering period. Increased K supply lowered THC and CBG in the inflorescence, raising P produced inconsistent effects, and humic acid mainly changed the mineral profile rather than the cannabinoids. The authors are careful with their conclusion: the relationship between cannabinoid content and nutrient supplementation is complex and non-linear. Anyone who keeps feeding beyond the optimum shifts the profile, usually not for the better. Caplan, Dixon and Zheng (2019) add a complementary perspective: controlled drought stress in late flower increased cannabinoid content, with no meaningful reduction in floral biomass reported. Water and nutrient regimes are therefore directly coupled; if you optimize fertilization, think the irrigation schedule through with it. Interactions that forums often leave out Four interactions belong on the radar of CSCs and growers. First, photoperiod: Saloner and Bernstein (2020) tested explicitly under 18/6, that is in veg. Bevan et al. (2021) tested at 12/12, that is in flower. The N optima of the two studies do overlap (160 to 194 mg/L), but the underlying logic differs: in veg, N feeds mass growth; in flower, it competes with cannabinoid metabolism. A feeding schedule that ignores the photoperiod switch leaves either yield or quality on the table. Second, genotype: the K study shows it most clearly, but it applies to N and P just as much. When you bring in a new cultivar, treat recommendations from studies on other cultivars as a starting value, not a target value, and watch the plants. Third, growing medium: Bevan et al. (2021) pointed to Caplan et al. (2017), who found a higher N optimum (212 to 261 mg/L) for organically based substrates. The explanation is mundane but important: organic N sources are not immediately plant-available. The nominal dose has to be higher to bring effective availability up to the level of a mineral solution. That is also the core of the review by Ahmadi et al. (2024) on chemical versus organic fertilization systems. That review mainly covers industrial hemp, but the mechanisms transfer to drug-type cannabis. Malík and Tlustoš (2025) and Nemati et al. (2021) add current reviews on growing media. Fourth, the microbiome. Ahmed and Hijri (2021) review the state of knowledge on mycorrhizae, PGPR and endophytes in cannabis. Their verdict is sobering: plausibly effective, barely established causally. And commercial “myco” products often do not contain the species that actually work in cannabis. If you experiment with biostimulants, treat it as a trial and not as a standard recommendation. Methods: what the studies deliver and what they do not Three limitations to keep in mind. First: small sample sizes. The Bernstein and Saloner studies work with n

Lead: 12 hours of light, 12 hours of darkness. For decades this rule has been the standard for bringing cannabis into flower indoors. Three studies from the last two years suggest that the standard protocol is suboptimal in many cases, and that the resulting yield loss can be measured in double-digit percentages. What this is about In the previous post we covered the broad lighting toolkit for cannabis cultivation (PPFD, spectrum, UV). One variable we deliberately only touched on: the photoperiod. It deserves its own post because the data here has become considerably stronger over the last two years than for many other cultivation levers. We looked at four studies: three direct photoperiod experiments on indoor cannabis (Peterswald et al. 2023, Ahrens et al. 2023, Ahrens et al. 2024), plus one nutrient study under a long photoperiod (Saloner and Bernstein 2020) as context for the vegetative phase. Three core findings that growers can act on right away. Finding 1: 14 hours of light means more flower, almost always In 2023, Peterswald and colleagues tested three cannabis cultivars against nine different photoperiods in a facility in Australia licensed by the Office of Drug Control (ODC). One CBD line (“Cannatonic”) and two THC lines (“Northern Lights”, “Hindu Kush”). They compared 10L:14D, 12L:12D and 14L:10D, plus switch treatments in mid flower (28 days after onset). The result was consistent across all three cultivars: flower yield was highest in treatments that started with 14L:10D. The plants received more light for photosynthesis in the early flowering phase, grew more vigorously and set more flower. A year later, Ahrens and colleagues from Guelph (Canada) replicated this with two THC-dominant cultivars (“Incredible Milk”, “Gorilla Glue”), this time with only 13 instead of 14 hours of light, but running the plants all the way to commercial maturity. Yield in the 13-hour treatment: 1.35 times the 12-hour yield in IM, 1.50 times in GG. That is 35 to 50% more flower. Worth noting: the DLI (daily light integral, the total amount of light per day) rose by only about 8% from 12 to 13 hours. The yield jump was therefore four to six times larger than the light quantity alone would explain. So the effect does not seem to be just “more light = more flower” but also a different physiological response of the plant. Finding 2: But THC can suffer, a cultivar question This is where it gets practical. In the two THC lines, Peterswald did measure more flower under 14L, but at the same time saw a significant drop in THC concentration. The trichomes were less dense, the leaves longer and narrower: the reproductive phase apparently had not run its full course. Cannatonic, the CBD line, showed the exact opposite: CBD concentration rose under the long photoperiod, and combined with the higher yield that added up to +50 to +100% total CBD per plant. For CBD programs, that is a substantial win. The THC lines, in contrast: higher flower mass but lower THC%. Whether you end up with more total THC per plant depends on how much the drop in concentration offsets the yield gain. In Northern Lights, the treatment that started at 14L and switched back to 12L delivered +50% yield without the THC dip. Here the Ahrens 2024 study partly disagrees: at 13 h (instead of 14 h), THC concentrations showed no significant difference or were even higher than under 12 h. Total THC per plant rose by at least 38%. In other words: 13 hours seems to be a kind of sweet spot that captures the yield advantage without stalling THC ripening. 14 hours goes one step too far in some cultivars. The takeaway: test cultivar by cultivar, in 0.5-hour steps between 12 and 14 h. The data does not point to one universal recommended value; it suggests that every cultivar has its own optimum. Finding 3: Some cultivars show no flowering delay at 14 h at all In a broader 2023 study, Ahrens et al. tested ten cultivars against six photoperiods between 12 and 15 hours of light. The study only tracked the early flowering phase (three to four weeks, roughly one third of the commercial ripening time). The picture is heterogeneous, as expected: Four cultivars (study codes BD, BT, IM, GJ) had their optimum between 12.4 and 12.7 hours, close to the 12-hour standard but slightly longer. Four cultivars (GT, LL, GG, OG) showed no delay at all in flowering onset up to 14 hours of light. Two cultivars (CC, PD) showed a linear delay, but only 1.3 and 2.3 days respectively at 13 h. Three cultivars even initiated flowering under 15 h, but the flower tissue did not develop further, so no practical yield advantage there. The bottom line for practice: photoperiods between 12 and 14 hours are worth testing. Beyond 14 hours, results get unstable. The optima apparently sit between 12.5 and 13 h for the early phase and can go up to 14 h if you wait for commercial maturity. What the studies do not settle yet Four points that remain open and should be considered before recommending “13 instead of 12”: Maturity drift at 13 h. In the Ahrens 2024 study, the 13-hour group was harvested when the 12-hour group was ripe. Clues such as slightly elevated CBGA concentrations suggest the 13-hour plants may not have reached their ripeness peak yet. If you run 13 h, decide harvest time by maturity indicators (stigma browning, trichome ambering), not by day count. Interaction with PPFD. All studies worked with moderate PPFD values (360 to 700 µmol m⁻² s⁻¹). At very high values (>900 µmol) the effects are unclear. Plausible: under very bright light, 12 h of DLI could already be saturating, so extra hours would add less. Cultivar library. Across the three studies, around 15 cultivars have been tested in total. That is far more than two years ago, but the pool of cannabis genetics is orders of magnitude broader. Generalizations to all cultivars should be treated with caution.

Every grow room has a thermo-hygrometer these days, and most have a VPD chart pinned to the wall somewhere. But when a member asks tomorrow why you run 50 % humidity in bloom, what do you say? “It says so on the internet” is not a good answer. Let’s look at what the research actually gives us. VPD for cannabis has become the grow scene’s favourite topic. VPD, vapour pressure deficit, bundles temperature and relative humidity into a single number and describes how strongly the air pulls water “out of the leaves”. That is handy, because it lets you steer the climate without juggling temperature and humidity separately in your head. But here is the catch: VPD for cannabis is far less well researched than the neatly gridded charts on grow room walls suggest. Exactly one direct cannabis study sits behind the topic, everything else is transfer from better-studied plants or industry consensus. For a club that wants to grow reproducibly and traceably, an honest look at what is proven, and what only sounds good, is worth it. What VPD Is Actually About In short: VPD describes how much water the air can still pull out of the plant. It is given in kilopascals (kPa) and rises as the air gets warmer or drier. The plant opens its stomata to take in CO₂ for photosynthesis. In doing so it loses water through the same openings. How much depends on the vapour pressure deficit: if the air is dry and warm (high VPD), a lot evaporates; if it is humid and cool (low VPD), little does. Both extremes cost yield. At too high a VPD the plant closes its stomata to avoid drying out, and thereby also shuts down CO₂ uptake, which slows photosynthesis. At too low a VPD, moisture builds up, transpiration as the “engine” for nutrient transport from the roots weakens, and the fungal risk rises. You can calculate VPD yourself, the formula is simple and built into every climate controller: from the leaf temperature (roughly air temperature) you get the saturation vapour pressure, from which you subtract the actual vapour pressure that follows from relative humidity. The result in kilopascals (kPa) is the VPD. Important in practice: because temperature is part of it, “60 % humidity” at 22 °C is a completely different VPD than at 28 °C. That is exactly why switching to a VPD value instead of pure humidity targets pays off. VPD for Cannabis: What Is Proven and What Is Transferred Here comes the uncomfortable part. If you go through the cannabis literature systematically (we did this in our own setpoint review), you find surprisingly little direct VPD research. The only clean cannabis VPD study comes from Sheldon et al. (2021): they tested 13 hemp varieties in a walk-in climate chamber against four VPD levels and found that the threshold at which the stomata close is variety-dependent. In roughly 5 of 13 varieties they close much earlier. Two catches for us: these are fiber hemp trials, not THC-rich genetics, and the study sits behind a paywall. Fiber hemp and medical cannabis differ enough in leaf anatomy and stomata architecture that you should not take absolute threshold values one to one. The second solid cannabis pillar is older: Chandra et al. (2008) measured photosynthesis response curves against temperature, CO₂ and light. That is the source of the often-cited finding that photosynthesis reaches its optimum at around 25 to 30 °C and saturates at roughly 750 ppm CO₂. It is not a VPD study, but it marks out the temperature corridor you should even be moving in. The rest of the VPD “wisdom” is transferred from better-researched C3 plants: tomato, lettuce, cucumber. Like these, cannabis is a C3 plant, so the mechanisms transfer, but the absolute numbers do not. The most important pieces carried over from C3 model plants: Setpoint systematics: Shamshiri et al. (2018) provide evidence-based T, RH and VPD ranges per growth stage for tomato, neatly graded into “optimal / borderline / critical”. Exactly this kind of synthesis is completely missing for cannabis. The framework carries over, but you have to adapt the actual values yourself. The stomata mechanism: Amitrano et al. (2021) showed in lettuce that low VPD leads to more and smaller stomata and raises net photosynthesis by around 18 %. The mechanism holds across C3 plants. The upper limit: Zhong et al. (2023) quantified a VPD threshold at ecosystem level (around 3.5 to 4.0 hPa) above which photosynthesis begins to tip. Together with Sheldon’s hemp thresholds, that gives a plausible basis for arguing an upper VPD limit. The absolute number from an ecosystem study is not, however, directly transferable to the grow room. The most exciting open field: Amitrano et al. (2021, second paper) found in lettuce that under different VPD (0.69 vs. 1.76 kPa) the profile of bioactive compounds shifts. The obvious hypothesis, that VPD could also influence cannabinoid and terpene profiles, is simply not yet tested for cannabis. Practical Setpoints, With an Honest Evidence Label The following tiering sums up what VPD for cannabis looks like once you put cannabis studies, C3 transfer and industry consensus side by side. Read the last column too: it tells you how firm the ground is under each value. “Proven” means direct cannabis evidence, “transferred” means derived from C3 plants, “consensus” means practical experience without a clean study. Stage Temperature (day) rel. humidity VPD (day) Confidence Clones / rooting 22 to 25 °C 70 to 80 % 0.4 to 0.7 kPa transferred Vegetative 24 to 28 °C 60 to 70 % 0.8 to 1.2 kPa partly proven (T), rest transferred Early bloom 24 to 26 °C 55 to 65 % 1.0 to 1.3 kPa consensus Late bloom 22 to 24 °C 45 to 55 % 1.2 to 1.6 kPa humidity from fungal risk, VPD transferred A few things that really stand out from the studies and that you can pass on to a member: The temperature corridor is the best proven. Chandra et

In the 1970s, botanists could not even agree on whether cannabis was one species or several. Today there are plants with three times as many chromosome sets as the original, and, in a first study, up to 65 percent more total cannabinoids. How we got here is a short history of plant breeding in fast-forward. The Start: A Plant Nobody Really Knew In 1976, two botanists drew a line under a 250-year-old debate. Ernest Small and Arthur Cronquist published a paper in the journal Taxon with the plain title “A practical and natural taxonomy for Cannabis” (Small & Cronquist, 1976). Their conclusion: there is only one biological species, Cannabis sativa L., with two subspecies and a wild and a cultivated variety each. The labels “Sativa” and “Indica” marketed today are, taxonomically, more of a marketing construct than a botanical finding, a view that contemporary reviews confirm (McPartland, 2018). What existed as breeding at that time ran largely underground. In California and the Netherlands, hobby breeders crossed landraces (regional, naturally adapted original strains) from Afghanistan, Mexico and Thailand and named the results Skunk #1, Northern Lights or Haze. Scientific genetics happened in parallel elsewhere: on fiber hemp in Wageningen, on industrial hemp in Bologna. The Mendel Turn: One Locus Governs the THC-CBD Ratio In 2003, Etienne de Meijer and his Italian team delivered one of the most important findings in cannabis genetics. In the journal Genetics, they showed that the ratio of THC to CBD in a plant is controlled by a single locus with two co-dominant alleles (de Meijer et al., 2003). That sounds technical, but in plain terms it means: cross a THC-dominant with a CBD-dominant plant and you can predict the cannabinoid profile of the offspring like Mendel’s peas. That opened the door to targeted variety development. Not by chance, a first wave of CBD-dominant hemp varieties appeared on the market in the years that followed, and the idea of cleanly defining “chemical varieties” or chemotypes became the standard (Russo, 2019). 2011: Cannabis Gets a Genome The next big leap came out of Toronto. Harm van Bakel and colleagues sequenced the genome of the drug strain Purple Kush in 2011 and compared it with the fiber hemp Finola (van Bakel et al., 2011). The result: around 534 megabases, 30,000 genes and, at the molecular level, clear differences in the cannabinoid biosynthesis pathways. Four years later, a team in Canada backed this up with marker data: hemp and marijuana are genetically clearly separate populations, but variety names often correlate surprisingly poorly with the underlying genetics (Sawler et al., 2015). Then, in 2018, the keystone: a high-resolution map showed that the genes for THC and CBD synthase sit in extremely rearranged genome regions, packed with jumping elements (Laverty et al., 2019). That explains why cannabinoid profiles can vary so strongly between varieties, and it provides the basis for modern marker-assisted selection. The fiber hemp field kept pace in parallel with its own reviews and Wageningen programs (Salentijn et al., 2014). Domestication: Older and More East Asian Than Thought For a long time, Central Asia counted as the cradle of cannabis. In 2021, Guangpeng Ren and colleagues presented a resequencing of 110 accessions worldwide in Science Advances and shifted the story eastward (Ren et al., 2021). Cannabis was, accordingly, domesticated around 12,000 years ago in East Asia; the sativa– and indica-type lines known today are the result of later selection. The original wild forms are largely extinct, which makes highland landraces from Afghanistan and the Hindu Kush a genetic treasure (McPartland & Small, 2020). If you want the full picture of genome research, you will find the roadmap in a review article: available assemblies, GWAS studies, sex determination, cannabinoid clusters (Hurgobin et al., 2020). The Cannabis sativa of 2026 is genetically better understood than many classic crops. Polyploids: The Latest Wave That leaves the current trend, and it is old and new at once. Polyploidy means a plant has three (triploid), four (tetraploid) or even six (hexaploid) chromosome sets instead of the usual two. This has worked in many crops for decades; in cannabis there were attempts early on, but it stayed a niche (Parsons et al., 2019). Triploids have a practical charm: they are largely sterile and barely form seeds. For flower production that is a real advantage: no unwanted pollination and more even flower mass. A controlled study from Connecticut confirmed in 2024 that triploid cannabis plants keep pace with diploids in growth and flower yield, but their seed formation is dramatically reduced (Kurtz et al., 2024). And then comes the latest paper, fresh from Korea: Tae Hyun Ha and his team presented the first stable hexaploid cannabis plants in early 2026 (Ha et al., 2026). With an optimized colchicine treatment and multi-generation selfing, they arrived at plants whose hexaploid status is cleanly confirmed by flow cytometry and chromosome counting. The study reports roughly 65 percent higher total cannabinoid content, about 60 percent more plant height and 2.7 times the dry flower mass compared with diploids. Whether this holds up in practice is for follow-up research to show, but the trend is clear. What Has Changed in 50 Years The plant is the same, what we know about it is not. In 1976 there was an argument over whether cannabis is one species. In 2026, cannabinoid profiles can be dialed in by marker selection, the domestication history reconstructed and varieties bred with triple the chromosome count. What remains open: most variety names on the market still correlate poorly with the actual genetics, and the diversity of the original landraces is under threat (Andre et al., 2016). So keep an eye on the next wave of cannabis research. It will probably not come from Wageningen or Toronto, but from where polyploids, domestication genetics and variety identity meet. Take-Aways One species, many chemotypes: cannabis is botanically a single species; the THC-CBD ratio depends on a single locus. Genome since 2011, map since 2018: modern cannabis breeding is marker-assisted and no