High light intensities can be used to grow healthy and robust cannabis plants during
the vegetative stage of indoor production
Melissa Moher1, David Llewellyn1, Max Jones2 and Youbin Zheng1,*
1School of Environmental Sciences and 2Department of Plant Agriculture, University of Guelph,
50 Stone Road East, Guelph, ON, N1G 2W1, Canada
*Corresponding author. E-mail address: firstname.lastname@example.org
We thank Ontario Centres of Excellence and HEXO Corp. for financial support and HEXO
Corp. for providing the plant material and production space for this experiment. We also thank
Bluelab Corporation for their measurement tools and to Scott Golem, Elizabeth Foley, Steve
Dinka, and Allison Slater for their outstanding technical support during the trial.
Additional index words. Light-emitting diodes, PPFD, DLI, growth, morphology
Although the vegetative stage of indoor cannabis production can be relatively short in duration,
there is a high energy demand due to higher light intensities (LI) than the clonal propagation
stage and longer photoperiods than the flowering stage (i.e., 16 – 24 hours vs. 12 hours). While
electric lighting is a major component of both energy consumption and overall production costs,
there is a lack of scientific information to guide cultivators in selecting a LI that corresponds to
their vegetative stage production strategies. To determine the vegetative plant responses to LI,
clonal plants of ‘Gelato’ were grown for 21 days with canopy-level photosynthetic photon flux
densities (PPFD) ranging between 135 and 1430 µmol·m-2·s-1 on a 16-hour photoperiod (i.e.,
daily light integrals of ≈ 8 to 80 mol·m-2·d-1). Plant height and growth index responded
quadratically; the number of nodes, stem thickness, and aboveground dry weight increased
asymptotically; and internode length and water content of aboveground tissues decreased linearly
with increasing LI. Foliar attributes had varying responses to LI. Chlorophyll content index
increased asymptotically, leaf size decreased linearly and specific leaf weight increased linearly
with increasing LI. Generally, PPFD levels of ≈ 900 µmol·m-2·s-1 produced compact, robust
plants that are commercially relevant, while PPFD levels of ≈ 600 µmol·m-2·s-1 promoted plant
morphology with more open architecture – to increase airflow and reduce the potential foliar
pests in compact (i.e., indica-dominant) genotypes.
Drug-type cannabis is a high-value crop that is mainly grown in controlled environments [e.g.,
indoors (i.e., with no natural lighting) and greenhouses] where growing conditions can be
maintained for consistent, year-round production (Benke and Tomkins, 2017; Despommier,
2013). Electricity costs are particularly high in indoor environments (Mills, 2012) because the
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 15 April 2021 doi:10.20944/preprints202104.0417.v1
© 2021 by the author(s). Distributed under a Creative Commons CC BY license.
plants completely rely on electric light sources for providing photosynthetically active radiation
(PAR, 400-700 nm). Electric lighting is also used in greenhouse environments to provide
supplemental PAR when the natural light levels are insufficient [e.g., when daylengths are short
or when it is cloudy outside (Bilodeau et al., 2019)]. Since light has a major role in moderating
plant morphology and ontogeny, light intensity (LI), spectrum, and photoperiod can be
manipulated by the cultivator to produce plants with the desired morphological characteristics
during the various growth stages of indoor cannabis production; ultimately resulting in high yield
and quality of the marketable products (e.g., mature female inflorescences). Lighting-related
electricity consumption is also a major consideration, due to its exceptionally high cost (e.g., per
unit of crop yield) in indoor cannabis production (Arnold, 2013; Mehboob et al., 2020).
Each of three distinct growth stages that are commonly used in indoor cannabis production (i.e.,
propagation, vegetative growth, and flowering) have different photoperiod and LI requirements.
In the propagation stage, the photoperiod is generally 18 – 24 h (Chandra et al., 2020) and
canopy-level photosynthetic photon flux density (PPFD, µmol·m-2·s-1) is usually low (Fluence,
2020; Lumigrow, 2017) to minimize transpiration loss as the clonal plants establish new root
systems. After approximately two weeks in propagation, rooted cuttings (i.e., transplants)
transition into the vegetative stage (Caplan et al. 2018) where they are exposed to similar
photoperiods but higher PPFD than propagation to encourage strong vegetative growth to
prepare the plants for the flowering stage (Rodriguez-Morrison et al, 2021). After approximately
two to four weeks in the vegetative stage, plants are transitioned to a 12-h photoperiod and even
higher PPFD to enhance growth and yield. Depending on the genotype, indoor-grown cannabis
crops normally spend between 6 and 12 weeks under the 12-h flowering photoperiod before the
female inflorescences have reached optimum maturity for harvesting (Carpentier et al., 2012).
The optimum post-vegetative stage morphology varies depending on the cultivators’ production
system (e.g., length of vegetative stage, plant density in both vegetative and flowering stages,
growing media type and rootzone volume, type of trellising system used in flowering, etc.), but
the general goal is to ensure high transplant success and strong vegetative growth (Vanhove et
al., 2011). The LI during the vegetative stage can influence plant growth attributes such as
height, stem thickness, branching, leaf size, leaf thickness, and biomass partitioning (Poorter et
al., 2019). Since these attributes affect a crop’s robustness as it enters the flowering stage, the
vegetative stage LI must be selected to promote the development of the foundational structure
(e.g., thicker stems and more nodes) needed to support prolific inflorescence development, which
can account for more than half of the total aboveground biomass at peak maturity (Rodriguez-
Morrison et al., 2021).
The current lack of scientific information related to LI during the vegetative stage has resulted in
a broad range of canopy-level PPFDs (e.g., 250 to 650 µmol·m-2·s-1) being recommended to
cultivators (Fluence, 2020; Lumigrow, 2017). Since cannabis can tolerate (Chandra et al., 2008)
and even flourish (Rodriguez-Morrison et al., 2021) under very high LI, there is opportunity to
elevate PPFD during the vegetative stage to enhance plant structure and shorten the length of the
vegetative stage. Therefore, the objective of this study was to determine the effects of a broad
range of LI on vegetative stage cannabis morphology and growth attributes, to guide cultivators
towards optimizing the LI for their specific production strategies.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 15 April 2021 doi:10.20944/preprints202104.0417.v1
Materials and Methods
Plant propagation and cultivation. Uniform clonal cuttings of the cannabis genotype ‘Gelato-29’
(short and bushy growth habit) were coated with 0.1% indole-3-butyric acid rooting hormone
(StimRoot #1; Master Plant-Prod Inc., Brampton, ON, Canada) at the base of each cutting and
inserted into cylindrical rockwool plugs (3.6 cm diameter × 4.0 cm height; Grodan, Milton, ON,
Canada) at one cutting per plug. Plugs were pre-soaked in a preventative biological fungicide
solution (RootShield WP; Bioworks, Victor, NY, USA) at 0.45 g·L-1 in distilled water, with a
final electrical conductivity (EC) of 0.7 dS·m-1 and pH of 5.2. The plugs were placed in
propagation trays (0.5 × 0.3 m, 50 Plug Pre-filled; A.M.A Horticulture Inc., Kingsville, ON,
Canada) and covered with transparent plastic lids (0.29 × 0.55 × 0.19 m, 7-inch Propagation
Dome; Mondi Products, Vancouver, BC, Canada). Cuttings were rooted for 14 d under a 16-h
photoperiod with a targeted canopy-level PPFD of 200 µmol·m-2·s-1 from light-emitting diodes
(LEDs) (Toplight-Targeted Spectrum; Lumigrow, Emeryville, CA, USA). Only the blue (B, 400-
500 nm) and red (R, 600-700 nm) channels were used, with peak wavelengths and full-width at
half maximum (FWHM) of 445 nm and 17 nm for red and 665 nm and 16 nm for blue, and a
photon flux ratio of B15:R85 (Fig. 1). Spectrum and LI were evaluated using a radiometrically-
calibrated spectrometer (XR-Flame-S; Ocean Optics, Dunedin, FL, USA) coupled to a CC3
cosine-corrector attached to a 1.9 m × 400 µm UV-Vis optical fibre. The intensities of B and R
LEDs were modified using the lighting control software (smartPAR; Lumigrow) to achieve the
prescribed PPFD and B:R.
Figure 1. Relative spectral photon flux distribution of blue (B) and red (R) LEDs used during the
propagation and vegetative stages.
Uniformly-sized rooted cuttings with height and number of nodes of (mean ± SE, n = 90) 13 ±
0.2 cm and 5 ± 0.1, respectively, were transplanted into rockwool blocks (0.15 × 0.15 × 0.15 m,
Grodan) and grown for 21 d under a 16-h photoperiod. The initial height, measured from
substrate surface to the highest point on the plant, and the number of nodes for each plant were
recorded. The transplants were not irrigated for the first three days to encourage root growth and
were then drip-irrigated twice daily at 2 L·hr-1 for 540 s, such that each plant received roughly
0.6 L·d-1. The nutrient solution was comprised of Dutch Nutrients Gro A and Gro B
Relative spectral photon flux
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 15 April 2021 doi:10.20944/preprints202104.0417.v1
(Homegrown Hydroponics, Toronto, ON, Canada) at a rate of 5 mL·L-1 in rain water, resulting in
an EC of ≈ 1.8 dS·m-1 and pH of ≈ 5.7.
The experiment was conducted in a commercial cannabis greenhouse facility in Southern
Ontario, Canada. Three enclosures (5.9 × 4.1 × 2.7 m) were used, each consisting of two benches
(5.9 × 1.8 m) that were separated by 0.5 m and encompassed with panda film (Vivosun, City of
Industry, CA, USA) – black side facing inwards – to block natural light and minimize solar
heating. Each enclosure was divided into five × 1 m2 plots, with a minimum lateral separation of
0.65 m between the edges of adjacent plots. Each plot consisted of 12 plants (i.e., 12 plants/m2),
arranged in four rows of three plants each, such that all plants were equally spaced. The plants in
the outer rows were border plants while the six plants in the inner rows were measured
experimentally (i.e., treatment plants). Plants were irrigated using the same nutrient solution that
was used during the transplant stage (described above). Air temperature and relative humidity
(RH) were recorded every 300 s using data loggers (HOBO MX2301A; Onset Computer
Corporation, Bourne, MA, USA) located at light fixture level in each enclosure. Across the three
enclosures, the daytime temperature and RH were (mean ± SD, n = 3) 25 ± 0.3 oC and 37 ± 0.6%
[i.e., vapor pressure deficit (VPD) ≈ 2.0 kPa], respectively, and nighttime temperature and RH
were 22 ± 0.1 oC and 40 ± 0.6% (i.e., VPD ≈ 1.6 kPa), respectively.
Light intensity treatments. This experiment was arranged as a randomized complete block design
(RCBD) with five target LI treatments (200, 450, 700, 950, and 1200 µmol·m-2·s-1) using the
same light fixtures and spectrum from the propagation stage (described above) and three
concurrent replications (i.e., the enclosures). Pairs of LED bars (1.09 × 0.11 m) were spaced 0.4
m apart ‘on-center’ over each plot. For the 1200 µmol·m-2·s-1 treatment plots, an additional pair
of LED bars were evenly spaced between the first pair of LED bars. All treatments had a
photoperiod of 16 h (0600 HR to 2200 HR). Spectrum and PPFD, at initial canopy level, were set
(as described above) using smartPAR (Lumigrow) and the spectrometer (Ocean Optics).
Following initial setup, the PPFD at the top of each plant was measured and recorded twice
weekly using a quantum sensor (LI-180; LI-COR Biosciences, Lincoln, NE, USA), and the
fixture hang-heights were adjusted accordingly, to maintain consistent canopy-level PPFDs
throughout the trial.
Although the layout of the experiment was a RCBD, the trial was conducted as a gradient design
(Jones-Baumgardt et al., 2020; Rodriguez-Morrison et al., 2021) with each plant treated as an
experimental unit and assigned a LI level consistent with their respective accumulated light
histories. To this end, the average PPFD (APPFD) each individual plant received over the trial
was obtained by computing the light integrals between each bi-weekly PPFD measurement
period, summing these integrals over the entire trial to determine a total light integral (TLI,
mol·m-2), and then back-calculating to determine APPFD by dividing TLI by the total number of
seconds of lighting during the trial (i.e., 3600 s·hr-1 × 16 hr·d-1 × 21 d).
Plant growth and leaf morphology. The plants were harvested 21 d after the start of the LI
treatments. Final height and number of nodes for each plant were recorded. Increases in height
(ΔH) and number of nodes (ΔNN) were determined by subtracting initial values from harvest
values. Internode length (IL) was determined by dividing ΔH by ΔNN. The width of each plant
was measured as the maximum lateral spread in two perpendicular axes based on the geographic
orientation on the bench: north-south (N-S) and east-west (E-W). Growth index (GI) was
calculated using the following equation: [(Final height × WidthN-S × WidthE-W) / 300] (from
Ruter, 1992). Chlorophyll content index (i.e., SPAD) was measured three times (then averaged)
on one of the youngest fully-expanded leaves using a chlorophyll meter (SPAD 502; Spectrum
Technologies Inc., Aurora, IL, USA). Stem thickness (ST) was measured at the first internode
using a digital caliper. The stem of each plant was cut at substrate level and aboveground fresh
weight (FW) was measured using a digital scale (AX622N/E Adventure Precision Balance;
OHAUS Corporation, Parsippany, NJ, USA). All aboveground tissues were dried to constant
weight at 65 oC and re-weighed to determine dry weight (DW). Aboveground tissue water
content (WC) was calculated using the following equation: [((FW - DW) / FW) × 100%]. Single
leaves from the tenth node from the bottom of each plant were scanned (Canoscan LiDE 25;
Canon Inc., Japan) at 600 dpi resolution and then dried to constant weight at 65 oC. Leaf size
(cm2/leaf) was computed from the digital images using ImageJ (Version 1.52q; National
Institutes of Health, Bethesda, MD, USA). The DW of each scanned leaf was determined using
an analytical balance (AE 100; Mettler Toledo, Columbus, OH, USA) and specific leaf weight
(SLW; mg·cm-2) was determined by dividing leaf DW by leaf size.
Data processing. All data were analyzed using least-squares non-linear regression in Prism
(GraphPad Software, San Diego, CA, USA) with APPFD as the independent variable, to
determine the best-fit model for each attribute (P ≤ 0.05). The models tested were linear,
quadratic, and asymptotic. Outliers were detected and removed using a Q-coefficient of 1.0 in
Prism’s ROUT outlier detection algorithm. For quadratic responses, the vertices were calculated
to determine the light saturation points (LSP) for each attribute. The asymptotic equation: Y = a
+ be(kX), where Y, a, e, and X represent the measured attribute, maximum value for the measured
attribute (i.e., the horizontal asymptote), Euler’s number, and APPFD, respectively, was used to
model non-linear relationships that did not have a vertex within the tested APPFD range. For
asymptotic models, maximum quantum efficiency (MQE) was derived from the slope of the
linear portion of the models, over the APPFD range of 130 to 200 µmol·m-2·s-1. Further, PPFD20
(i.e., a practical LSP) was defined for the asymptotic models as the APPFD level where the
localized slope of the curve fell below 20% of the slope at MQE. The PPFD20 was used to
indicate that increasing the APPFD beyond this level resulted in minimal further increases in the
respective responses; thus, acting as a proxy for a LI-response efficiency threshold.
The range of APPFDs that plants grew under in this trial was 135 to 1430 µmol·m-2·s-1,
corresponding to daily light integrals (DLI) ranging from 7.8 to 82 mol·m-2·d-1. Notably, there
were no signs of transplant shock or light stress, even in plants placed under the highest LIs
(which were up to 7 times higher than the LI in the propagation stage). Overall, plants grown
under different LIs exhibited varying architectures (Fig. 2) and leaf morphology (Fig. 3).
Generally, plants grown under high LI had more compact, denser growth, resulting in shorter
plants, greater numbers of potential flowering sites, and higher aboveground biomass. However,
individual measured growth attributes had varying responses to increasing LI. While some
attributes exhibited linear responses to LI, several attributes exhibited saturating responses to
increasing LI, and others had maxima at moderate APPFD levels.
Figure 2. Cannabis plants after growing under canopy-level average photosynthetic photon flux
densities (APPFD) of 179, 478, 713, 917, and 1367 µmol·m-2·s-1 with a 16-h photoperiod for 21
Figure 3. Single cannabis leaves taken at the tenth node after growing under canopy-level
average photosynthetic photon flux densities (APPFD) of 160, 477, 716, and 1043 µmol·m-2·s-1
with a 16-h photoperiod for 21 d.
Increasing LI resulted in smaller leaflets with smaller, more numerous serrations along the leaflet
margins (Fig. 3). Individual leaf size decreased linearly (Fig. 4A) and individual leaf biomass
increased linearly (data not shown) resulting in an 84% increase in SLW (Fig. 4B) at the
maximum vs. minimum APPFD. SPAD, an area-based index of chlorophyll content, increased
asymptotically with increasing LI, and was 24% higher at the PPFD20 of 1030 vs. 135 µmol·m-
2·s-1 (Fig. 4C). The ΔNN and ST also increased asymptotically with increasing LI, with
respective PPFD20 of 472 and 870 µmol·m-2·s-1, where ΔNN and ST were 28% and 41% higher
vs. the minimum APPFD (Fig. 4D and E). The IL decreased linearly with increasing LI, resulting
in 24% shorter internodes at the maximum vs. minimum APPFD (Fig. 4F). Both ΔH and GI (of
which final height is a coefficient) had quadratic responses to LI, with maxima at 686 and 582
µmol·m-2·s-1, respectively (Fig. 4G and H). The maximum ΔH was 12% and 24% higher than at
the minimum and maximum APPFD, respectively and the maximum GI was 14% and 76%
higher than at the minimum and maximum APPFD, respectively. Aboveground DW increased
asymptotically with increasing LI and was 2.6 times higher at the PPFD20 of 910 vs. the
minimum APPFD (Fig. 4I) while WC decreased linearly by 9% at maximum vs. minimum
APPFD (Fig. 4J).
Figure 4. Individual leaf area (A) and specific leaf weight of individual leaves taken at the tenth
node (B), leaf chlorophyll content index (i.e., SPAD value) of the youngest fully-expanded leaf
(C), increase in the number of nodes (D), stem thickness (E), internode length (F), increase in
height (G), growth index (H), aboveground dry weight (I), and aboveground plant tissue water
content (J) of vegetative cannabis plants grown for 21 d under average photosynthetic photon
flux densities (APPFD) ranging from 135 to 1430 µmol·m-2·s-1. Each data point represents an
individual plant with its own APPFD.
In the indoor cannabis production industry, there is considerable variability in the
characterization of what constitutes an optimum structure of clonal plants prior to the initiation
of the (flower-inducing) short-day photoperiod. This is due to myriad factors, including:
genotypic specific growth habit [e.g., indica- vs. sativa-dominant plant structure (Jin et al.,
2021)], size of plants, substrate volume, cropping density, environmental settings (including LI),
and many cultivator-specific plant husbandry practices such as periodic de-leafing and utilization
of plant training (e.g., stakes, trellis-supports, etc.). Notwithstanding these variances, the
underlying goals of the vegetative stage are steadfast: to produce healthy, resilient plants that are
capable of supporting prolific inflorescence biomass production, from both assimilative and
structural perspectives. Therefore, within the aforementioned cultivator-specific constraints,
plants coming out of the vegetative stage should have a general structure that is primed to
optimize future photosynthetic capacity, facilitate airflow within the crop canopy, maximize
potential flowering sites, and bear the weight of the mature inflorescences. These parameters
necessitate plants that have foliar architecture and morphology capable of intercepting and
utilizing the incoming PAR, with as many nodes as possible [cannabis flower buds arise from
foliar axils (Spitzer-Rimon et al., 2019)], and that have relatively compact growth (i.e., short
internodes) with robust stems.
Key plant morphological and physiological attributes have shown varying responses to LI. In a
comprehensive review paper, Poorter et al. (2019) summarized the characteristic responses of
many attributes from myriad herbaceous and woody plants using relative response models over
DLIs up to 50 mol·m-2·d-1 (i.e., equivalent to ≈ 870 µmol·m-2·s-1 in the present study).
Extrapolating their findings to the APPFD range in the present study, they found that individual
leaf area decreased by ≈ 23% and SLW nearly doubled with increasing LI, although there were
no LI treatment effects on area-based chlorophyll content. The LI treatment effects on leaf
morphology were somewhat smaller in Poorter et al. (2019) compared to the present study,
suggesting that cannabis may have relatively high phenotypic plasticity for leaf morphology
adaptations to LI. However, the present study observed a 24% increase in area-based chlorophyll
content, which may indicate that cannabis favours upregulating photosynthetic capacity (i.e.,
maximizing resource utilization) over the common foliar morphology-based adaptive responses
to high light stress. Clonal cannabis’ very high photosynthetic capacity (Chandra et al., 2008;
Rodriguez-Morrison et al., 2021) appears to be present even at the relatively young vegetative
stage (Chandra et al., 2015). In the context of indoor production, the reduction in individual leaf
area with increasing LI may also confer an increase in whole-plant net photosynthesis, since a
greater proportion of the incident PAR should penetrate deeper into the canopy through inherent
reductions in self-shading. Moreover, leaves with higher SLW, which is strongly correlated with
leaf thickness (Vile et al., 2005; Wilson et al., 1999), can increase water use efficiency (Yun and
Taylor, 1986), enhance resistance to pathogens (Guest and Brown, 1997), and minimize
The intensity of PAR in the vegetative stage can have major influences on plant structure during
this short but critical stage of production. Though not often reported (because it is a destructive
measurement), aboveground biomass (i.e., DW) is perhaps the single most comprehensive
parameter that relates LI effects on vegetative growth. As it does in floral and non-floral biomass
at optimum inflorescence maturity (Rodriguez-Morrison et al., 2021), DW during the vegetative
stage had a strong linear response to increasing LI. There was almost a 3-fold increase in DW
over the 135 to 1430 µmol·m-2·s-1 APPFD range in the present study, although 90% of the
maximum increase in DW was attained at an APPFD of only ≈ 900 µmol·m-2·s-1. Further,
aboveground tissue moisture content decreased linearly with increasing LI (Fig. 4J), which is a
common response to LI (Poorter et al., 2019) that normally confers an increase in mechanical
strength (Shah et al., 2017). Both ΔH and GI were maximized at moderate APPFD levels of ≈
600 µmol·m-2·s-1. While these are generally negative characteristics in the context of vegetative-
stage cannabis, open plant architectures may benefit denser genotypes (e.g., indica-dominant) by
increasing the airflow within the canopy, potentially suppressing foliar pests while making
routine pest monitoring easier (Bakro et al., 2018; Chandra et al., 2017). In contrast, plants were
smaller at ≈ 900 vs. 600 µmol·m-2·s-1 but had ≈ 15% higher DW and ≈ 6% thicker stems (i.e., ≈
13% higher cross-sectional area). Since the number of nodes saturated at relatively low LI, a
canopy-level PPFD target of about 900 µmol·m-2·s-1 may be most appropriate for producing
robust but not overly compact plants while also minimizing lighting-related energy and
infrastructure costs. Although not as common in commercial settings, production facilities that
target more open plant architecture and greater energy conservation may opt for canopy-level
PPFD target of ≈ 600 µmol·m-2·s-1.
Another consideration is the adaptive capacity of vegetative plants to the normal increases in
canopy-level LI as they transition into the flowering phase, which are necessary to maintain the
DLI in conjunction with shortening the photoperiod to induce strong flowering responses –
normally from ≥ 16 h to ≤ 12 h (Potter, 2014). Therefore, to maintain the same DLI as in the
vegetative phase, the PPFD must be increased by at least 25%. However, cannabis takes time to
acclimate its photosynthetic capacity to higher LIs when transitioning out of the vegetative phase
(Rodriguez-Morrison et al., 2021). Given vegetative cannabis’ demonstrated capacity to
proliferate under high LIs, using canopy-level PPFDs ≥ 900 µmol·m-2·s-1, particularly in the
latter stages of the vegetative phase (i.e., after plants have recovered from transplant shock), may
optimize their adaptation to the higher LIs in the flowering phase while also potentially
shortening the vegetative phase.
The industry recommendations for LI during cannabis’ vegetative stage are variable (e.g.,
Fluence, 2020; Lumigrow, 2017); however, few contemporary recommendations suggest
exposing vegetative cannabis plants to PPFDs higher than 800 µmol·m-2·s-1 in indoor production
systems. The current study demonstrates that vegetative cannabis can be exposed to substantially
higher LIs (than commonly-used in the industry) with positive morphological outcomes that can
prime plants for the transition into the flowering phase.
Within the parameters of this investigation, we observed that PPFD levels between 600 and 900
µmol·m-2·s-1 appeared to achieve an appropriate balance in optimizing key morphological
parameters in vegetative cannabis while minimizing energy use associated with excessively-high
LIs and also considering different production strategies. Although the desired morphological and
growth attributes of vegetative-stage clonal cannabis plants will be subjective to each genotype
and production scenario, the presented LI responses can assist cultivators in optimizing the LI for
their individual production goals; balancing the potential economic returns against elevated input
costs associated with supplying more PAR to their crops.
Arnold, J. M. 2013. Energy consumption and environmental impacts associated with Cannabis
cultivation. Humboldt State Univ., Arcata, MSc. Diss.
Bakro, F., W. Katarzyna, M. Bunalski, and M. Jedryczka. 2018. An overview of pathogen and
insect threats to fibre and oilseed hemp (Cannabis sativa L.) and methods for their biocontrol.
Integrated Control in Oilseed Crops 136:9–20.
Benke, K. and B. Tomkins. 2017. Future food-production systems: vertical farming and
controlled-environment agriculture. Sustainability: Sci. Practice Policy 13(1):13–26.
Bilodeau, S.E., B. Wu, A. Rufyikiri, S. MacPherson, and M. Lefsrud. 2019. An update on plant
photobiology and implications for cannabis production. Frontiers Plant Sci. 10:296.
Caplan, D., J, Stemeroff, M. Dixon, and Y. Zheng. 2018. Vegetative propagation of cannabis by
stem cuttings: effects of leaf number, cutting position, rooting hormone and removal of leaf tips.
Can. J. Plant Sci. 98(5): 1126-1132. https://doi.org/10.1139/cjps-2018-0038
Carpentier, C., K. Mulligan, L. Laniel, D. Potter, B. Hughes, L. Vandam, D. Olszewski, and K.
Skarupova. 2012. Botany and cultivation of cannabis, p. 20–39. In: C. Carpentier, L. Laniel, and
P. Griffiths (eds.). Cannabis production and markets in Europe. European Monitoring Centre for
Drugs and Drug Addiction, Lisbon, Portugal. https://doi.org/10.2810/76378
Chandra, S., H. Lata, I.A. Khan, and M.A. Elsohly. 2008. Photosynthetic response of Cannabis
sativa L. to variations in photosynthetic photon flux densities, temperature and
CO2 conditions. Physiol. Mol. Biol. Plants 14(4):299–306. https://doi.org/10.1007/s12298-008-
Chandra, S., H. Lata, I.A. Khan, and M.A. ElSohly. 2017. Cannabis sativa L.: botany and
horticulture, p. 79–100. In: S. Chandra, H. Lata, and M.A. ElSohly (eds.). Cannabis sativa L. –
botany and biotechnology. Springer Nature, Cham, Switzerland. https://doi.org/10.1007/978-3-
Chandra, S., H. Lata, and M.A. ElSohly. 2020. Propagation of cannabis for clinical research: An
approach towards a modern herbal medicinal products development. Frontiers Plant Sci. 11:958.
Chandra, S., H. Lata, Z. Mehmedic, I.A. Khan, and M.A. ElSohly. 2015. Light dependence of
photosynthesis and water vapor exchange characteristics in different high Δ9-THC yielding
varieties of Cannabis sativa L. J. Appl. Res. Medicinal Aromatic Plants 2(2):39-47.
Despommier, D. 2013. Farming up the city: The rise of urban vertical farms. Trends Biotechnol.
Fluence. 2020. Cannabis cultivation guide. Fluence, Austin, TX.
Guest, D. and J. Brown. 1997. Plant defences against pathogens, p. 263–286. In: J.F. Brown and
H.J. Ogle (eds.). Plant pathogens and plant diseases. Rockvale Publications, Armidale, Australia.
Jin, D., P. Henry, J. Shan, and J. Chen. 2021. Identification of phenotypic characteristics in three
chemotype categories in the genus Cannabis. HortScience 1-10.
Jones-Baumgardt, C., D. Llewellyn, and Y. Zheng. 2020. Different microgreen genotypes have
unique growth and yield responses to intensity of supplemental PAR from light-emitting diodes
during winter greenhouse production in Southern Ontario, Canada. HortScience 55(2):156-163.
Lumigrow. 2017. LED grower’s guide for cannabis. Lumigrow, Emeryville, CA.
Mehboob, N., H.E.Z. Farag, and A.M. Sawas. 2020. Energy consumption model for indoor
cannabis cultivation facility. J. Power Energy 7:222–233. https://10.1109/OAJPE.2020.3003540
Mills, E. 2012. The carbon footprint of indoor Cannabis production. Energy Policy 46:58–67.
Poorter, H., Ü. Niinemets, N. Ntagkas, A. Siebenkäs, M. Mäenpää, S. Matsubara, and T.L. Pons.
2019. A meta-analysis of plant responses to light intensity for 70 traits ranging from molecules to
whole plant performance. New Phytologist 223:1073–1105. https://doi.org/10.1111/nph.15754
Potter, D.J. 2014. A review of the cultivation and processing of cannabis (Cannabis sativa L.) for
production of prescription medicines in the UK. Drug Testing Analysis 6(1–2):31–38.
Rodriguez-Morrison, V., D. Llewellyn, and Y. Zheng. 2021. Cannabis yield, potency, and leaf
photosynthesis respond differently to increasing light levels in an indoor environment. Preprints.
Ruter, J.M. 1992. Influence of source, rate, and method of applicating controlled release fertilizer
on nutrient release and growth of ‘Savannah’ holly. Fertilizer Res. 32(1):101–106.
Shah, D.U., T.P. Reynolds, and M.H. Ramage. 2017. The strength of plants: theory and
experimental methods to measure the mechanical properties of stems. J. Expt. Bot. 68(16):4497-
Spitzer-Rimon, B., S. Duchin, N. Bernstein, and R. Kamenetsky. 2019. Architecture and
florogenesis in female Cannabis sativa plants. Frontiers Plant Sci. 10:350.
Vanhove, W., P. Van Damme, and N. Meert. 2011. Factors determining yield and quality of
illicit indoor cannabis (Cannabis spp.) production. Forensic Sci. Intl. 212(1–3):158–163.
Vile, D., E. Garnier, B. Shipley, G. Laurent, M.L. Navas, C. Roumet, S. Lavorel, S. Díaz, J.G.
Hodgson, F. Lloret, G.F. Midgley, H. Poorter, M.C. Rutherford, P.J. Wilson, and I.J. Wright.
2005. Specific leaf area and dry matter content estimate thickness in laminar leaves. Ann.
Bot. 96(6):1129-1136. https://doi.org/10.1093/aob/mci264
Wilson, P.J., K.E.N. Thompson, and J.G. Hodgson. 1999. Specific leaf area and leaf dry matter
content as alternative predictors of plant strategies. New Phytologist 143(1):155-162.
Yun, J.I. and S.E. Taylor. 1986. adaptive implications of leaf thickness for sun‐and shade‐grown
Abutilon theophrasti. Ecology 67(5):1314–1318. https://doi.org/10.2307/1938687