MANIPULATING PHOTON FLUX DENSITY, PHOTON SPECTRUM, AND
PHOTOPERIOD TO IMPROVE THE GREENHOUSE PRODUCTION OF SPECIALTY CUT
FLOWERS
By
Caleb Edward Spall
A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Horticulture−Master of Science
2022
ABSTRACT
MANIPULATING PHOTON FLUX DENSITY, PHOTON SPECTRUM, AND
PHOTOPERIOD TO IMPROVE THE GREENHOUSE PRODUCTION OF SPECIALTY CUT
FLOWERS
By
Caleb Edward Spall
Year-round demand for locally sourced specialty cut flowers continues to increase in the
U.S., but growers in northern regions cannot produce them outdoors or in high-tunnels year-
round due to poor environmental conditions during the winter and early spring. Thus, they must
use greenhouses to maintain a proper lighting environment to capitalize on demand during these
seasons. Our objectives for Expts. 1 and 2 were to a) determine how photoperiod influences
morphology of marigold ‘Xochi’ (Tagetes erecta), witchgrass ‘Frosted Explosion’ (Panicum
capillare), and dianthus ‘Amazon Rose Magic’ and ‘Amazon Neon Cherry’ (Dianthus barbatus
interspecific) young and finished plants, and b) quantify how photoperiod and daily light integral
(DLI) influence floral initiation and quality of witchgrass and dianthus. For Expt. 3, we sought to
quantify how supplemental radiation quality influences floral initiation and finished quality of
three long-day specialty cut flowers. Marigold was harvestable when young plants were grown
under photoperiods ≥11 h or a 4-h NI, and finished under 12-h photoperiods. Witchgrass was
harvestable when young plants were grown under photoperiods ≥14 h or a 4-h NI, and finished
under photoperiods ≥14 h. Dianthus was harvestable when young and finished plants were grown
under 9- to 16-h photoperiods, or a 4-h NI. Additionally, cut flowers should be grown under a
DLI of ≥10 mol·m–2·d–1. Godetia, snapdragon, and stock cut flowers should be grown under
supplemental light with a spectrum similar to B20R85FR15 or a broad spectrum, for desirable crop
quality and minimal developmental, visibility, and energy tradeoffs.
To Liz, Maisie, and Phoebe
iii
ACKNOWLEDGEMENTS
Throughout my graduate school journey, I have been guided and supported by my major
advisor, Dr. Roberto Lopez. Thank you for your patience, advice, and support. I am truly grateful
to have had you as a major professor. Additionally, I would like to thank Dr. Erik Runkle for
serving on my advisory committee. Your advice and writing suggestions were always welcome. I
would also like to thank to Dr. James Faust for serving on my advisory committee and providing
me with the opportunity to experience the floriculture industry abroad. Additionally, I would like
to thank Nate DuRussel for his greenhouse technical assistance.
I am grateful for support from the American Floral Endowment and the Association of
Specialty Cut Flower Growers throughout my graduate program. Having support from these
organizations allowed me to learn through travelling to grower operations and conferences and
afforded me opportunities to share my research at various symposiums. I would also like to thank
BloomStudios and Sakata Seed America for seeds and funding, Fluence Bioengineering,
Heliospectra, LumiGrow, P.L. Light Systems, and Signify for supplemental lighting fixtures,
Ludvig Svensson for shade cloth, Hydrofarm for netting, Syndicate Sales for floral supplies,
Blackmore for fertilizer, and Raker-Roberta’s Young Plants for sowing seeds.
I would like to thank Anthony Soster, Annika Kohler, Devin Brewer, Hyeonjeong Kang,
and Sean Tarr for their support, advice, writing suggestions, and friendship. I would also like to
thank John Gove, Alec Fowler, and Ian Holcomb for data collection assistance. Finally, I’d like
to thank my girlfriend Liz, and our two cats Maisie and Phoebe, for their endless support and
love throughout my graduate program.
iv
TABLE OF CONTENTS
LIST OF TABLES ...................................................................................................................... vii
LIST OF FIGURES ....................................................................................................................... x
SECTION 1 .................................................................................................................................. 1
Literature Review: Manipulating photon flux density, photon spectrum, and photoperiod to
improve the greenhouse production of specialty cut flowers ........................................................ 2
Introduction ............................................................................................................................. 2
High Tunnel Cut Flower Production ....................................................................................... 5
Controlled-Environment Greenhouse Cut Flower Production ................................................ 8
Photoperiodic Crop Lighting in Controlled-Environment Greenhouses ............................ 8
Radiation Intensity in Controlled-Environment Greenhouses ............................................15
Radiation Quality in Controlled-Environment Greenhouses..............................................19
Conclusion ...............................................................................................................................24
REFERENCES ..............................................................................................................................25
SECTION 2 ..................................................................................................................................35
Daily light integral and/or photoperiod during the young plant and finishing stages influence floral
initiation and quality of witchgrass and marigold cut flowers ......................................................36
Abstract....................................................................................................................................37
Introduction .............................................................................................................................39
Materials and methods .............................................................................................................43
Young plant material, culture, lighting treatments, and greenhouse environment.............43
Finished plant lighting treatments, greenhouse environment, and culture ........................45
Data collection and analysis ...............................................................................................43
Results .....................................................................................................................................47
Young plant morphology and dry mass...............................................................................47
Time to visible flower bud ...................................................................................................48
Node number below visible flower bud ...............................................................................49
Time to open flower of witchgrass ......................................................................................50
Witchgrass stem length, caliper, and branch number at open flower ................................50
Time to harvest ....................................................................................................................52
Marigold stem length, caliper, branch and inflorescence number at harvest ....................52
Discussion................................................................................................................................53
Acknowledgements .................................................................................................................58
APPENDIX ...................................................................................................................................59
REFERENCES ..............................................................................................................................74
SECTION 3 ..................................................................................................................................78
Daily light integral, but not photoperiod, commercially influences time to flower and finished
quality of dianthus specialty cut flowers .......................................................................................79
Abstract....................................................................................................................................80
v
Introduction .............................................................................................................................82
Materials and methods .............................................................................................................86
Young plant material, culture, lighting treatments, and greenhouse environment.............86
Finished plant lighting treatments, greenhouse environment, and culture ........................88
Data collection and analysis ...............................................................................................89
Results .....................................................................................................................................90
Young plant morphology, dry mass, and off-type incidence ...............................................90
Time to visible flower bud and node count .........................................................................91
Time to open flower.............................................................................................................92
Time to harvest ....................................................................................................................92
Cut flower morphology at harvest .....................................................................................93
Discussion................................................................................................................................95
Acknowledgements .................................................................................................................98
APPENDIX ................................................................................................................................. 100
REFERENCES ............................................................................................................................ 117
SECTION 4 ................................................................................................................................ 121
Supplemental lighting quality influences time to flower and finished quality of three long-day
specialty cut flowers .................................................................................................................... 122
Abstract.................................................................................................................................. 123
Introduction ........................................................................................................................... 125
Materials and methods ........................................................................................................... 131
Plant material, culture, and lighting treatments ............................................................... 131
Data collection and analysis ............................................................................................. 134
Results ................................................................................................................................... 135
Time to visible flower bud ................................................................................................. 135
Time to open flower........................................................................................................... 135
Time to harvest .................................................................................................................. 136
Cut flower morphology at harvest .................................................................................... 137
Flower petal coloration at harvest.................................................................................... 138
Discussion.............................................................................................................................. 138
Acknowledgements ............................................................................................................... 144
APPENDIX ................................................................................................................................. 145
REFERENCES ............................................................................................................................ 156
vi
LIST OF TABLES
Table 2.1. Actual average daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], air average
daily temperature (ADT), day temperature, and night temperature [mean ± SD (°C)] throughout
the duration of the witchgrass and marigold young-plant stage for reps. 1 and 2. ...................... 60
Table 2.2. Actual average daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], air average
daily temperature (ADT), mean day temperature, and mean night temperature [mean ± SD (°C)]
throughout the duration of the witchgrass finishing stage for reps. 1 and 2. ...............................61
Table 2.3. Actual daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], air average daily
temperatures (ADTs), mean day temperature, and mean night temperature [mean ± SD (°C)]
throughout the duration of the marigold finishing stage for reps. 1 and 2. ..................................62
Table 2.4. Effects of young-plant photoperiod (9, 12, 13, 14, 16, 18, 24 h, or a 4-h NI) and finishing
photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on stem caliper (mm) of witchgrass ‘Frosted
Explosion’ (Panicum capillare) at open flower. Cut flowers were finished under a moderate DLI
of ≈10 mol∙m–2∙d–1. . ......................................................................................................................63
Table 2.5. Effects of young-plant photoperiod (9, 12, 13, 14, 16, 18, 24 h, or a 4-h NI) and finishing
photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to harvest (d) from the date of
transplant of witchgrass ‘Frosted Explosion’ (Panicum capillare) grown under a moderate DLI of
≈10 mol∙m–2∙d–1. . ..........................................................................................................................64
Table 2.6. Effects of young-plant photoperiod (11, 13, 14, 15, 16, or 24 h, or a 4-h NI) and finishing
photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to harvest (d) from the date of
transplant of marigold ‘Xochi’ (Tagetes erecta). ......................................................................... 65
Table 2.7. Regression analysis equations and adjusted R2 for height; root dry mass; and shoot dry
mass in response to photoperiod (P; 9-, 11-, 12-, 13-, 14-, 15-, 16-, 18-, 24-h photoperiods or a 4-
h NI) of marigold ‘Xochi’ (Tagetes erecta) or witchgrass ‘Frosted Explosion’ (Panicum capillare).
All models are in the form of: ƒ = y0 + a*P + b*P2. ..................................................................... 66
Table 2.8. Regression analysis equations and adjusted R2 for time to visible flower bud; time to
open flower; and stem length at open flower in response to young-plant photoperiod (YP; 9, 12,
13, 14, 16, 18, or 24 h, or a 4-h night interruption; NI) and finishing photoperiod (FP; 10, 11, 12,
13, 15, or 16 h, or a 4-h NI) of witchgrass ‘Frosted Explosion’ (Panicum capillare). All models
are in the form of: ƒ = y0 + a*YP + b*FP + c*YP2 + d*FP2+e*(YP*FP). Cut flowers were finished
under a moderate DLI of ≈10 mol∙m–2∙d–1 or a very low DLI of ≈3 mol∙m– 2∙d– 1. ....................... 67
Table 2.9. Regression analysis equations and adjusted R2 for time to visible flower bud in response
to young-plant photoperiod (11, 13, 14, 15, 16, 24 h, or a 4-h NI) and/or finishing photoperiod (10,
11, 12, 13, 14, 15, 16 h, or a 4-h NI) of marigold ‘Xochi’ (Tagetes erecta). Models 3A is in the
vii
form of: ƒ = y0 + a*YP + b*FP + c*YP2 + d*FP2 and models 3B and 3C are in the form of: ƒ = y0
+ a*P + b*P2. ................................................................................................................................. 68
Table 3.1. Actual daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], air average daily
temperatures (ADTs), mean day temperatures, and mean night temperatures [mean ± SD (°C)]
throughout the duration of the young-plant stage for reps. 1 and 2. ........................................... 101
Table 3.2. Actual daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], air average daily
temperatures (ADTs), mean day temperatures, and mean night temperatures [mean ± SD (°C)]
throughout the duration of the finishing stage for reps. 1 and 2.................................................. 102
Table 3.3. Regression analysis equations and adjusted R2 for height; root dry mass; and shoot dry
mass in response to photoperiod (P; 9-, 11-, 12-, 13-, 14-, 15-, 16-, 18-, 24-h photoperiods or a 4-
h NI) of dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’ (Dianthus barbatus
interspecific) young plants grown under a moderate or very low DLI. All models are in the form
of: ƒ = y0 + a*P + b*P2. ............................................................................................................. 103
Table 3.4. Regression analysis equations and adjusted R2 for time to visible flower bud, time to
open flower, time to harvest, and stem length at harvest in response to young-plant photoperiod
(YP; 9, 11, 12, 13, 15, or 16 h) and/or finishing photoperiod (FP; 11, 12, 13, 14, 15, or 16 h, or a
4-h NI) of dianthus ‘Amazon Neon Cherry’ (Dianthus barbatus interspecific). All models are in
the form of: ƒ = a*YP + b*FP + c*YP2 + d*FP2+e*(YP*FP) unless otherwise indicated. ....... 104
Table 3.5. Regression analysis equations and adjusted R2 for time to visible flower bud, time to
open flower, time to harvest, and stem length at harvest in response to young-plant photoperiod
(YP; 9, 11, 12, 13, 15, or 16 h) and/or finishing photoperiod (FP; 11, 12, 13, 14, 15, or 16 h, or a
4-h NI) of dianthus ‘Amazon Rose Magic’ (Dianthus barbatus interspecific). All models are in
the form of: ƒ = a*YP + b*FP + c*YP2 + d*FP2+e*(YP*FP) unless otherwise indicated. ....... 105
Table 4.1. Actual daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], average daily
temperatures (ADTs), mean day temperature, mean night temperature, and mean leaf temperature
[mean ± SD (°C)] for each supplemental light (SL) treatment during the vegetative (VEG) and
reproductive (REP) stages of replication 1. SL treatments consisted of either 460-W HPS fixtures
(HPS120; LR48877; P.L. Light Systems, Beamsville, ON, Canada), 631-W LED fixtures
(B20G50R45FR5; VYPR 2p; Fluence, Austin, TX), 325-W LED fixtures (B20R85FR15; LumiGrow
Pro 325; LumiGrow, Emeryville, CA), 600-W LED fixtures (B30G25R65; LX601G, Heliospectra,
Göteborg, Sweden), a combination of 72-W LED fixtures (B120; HortiLED MULTI, P.L. Light
Systems) and 625-W LED fixtures (R120; LumiGrow Pro 650E, LumiGrow), or 625-W LED
fixtures (LumiGrow Pro 650E; LumiGrow)................................................................................ 146
Table 4.2. Actual daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], average daily
temperatures (ADTs), mean day temperature, mean night temperature, and mean leaf temperature
[mean ± SD (°C)] for each supplemental light (SL) treatment during the vegetative (VEG) and
reproductive (REP) stages of replication 2. SL treatments consisted of either 460-W high-pressure
sodium fixtures (HPS120; LR48877; P.L. Light Systems, Beamsville, ON, Canada), 631-W light-
emitting diode (LED) fixtures (B20G50R45FR5; VYPR 2p; Fluence, Austin, TX), 325-W LED
viii
fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville, CA), 600-W LED fixtures
(B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a combination of 72-W LED fixtures
(B120; HortiLED MULTI, P.L. Light Systems) and 625-W LED fixtures (R120; LumiGrow Pro
650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro 650E; LumiGrow). ............................
..................................................................................................................................................... 147
Table 4.3. Estimated phytochrome photoequilibria (PPE; PFR/PR+FR) and color fidelity index (CFI;
Rf) of each supplemental lighting (SL) treatment. PPEs were calculated according to Sager et al.
(1988) and CFI values were calculated according to supplemental materials provided by IES
(2018). SL treatments consisted of either 460-W HPS fixtures (HPS120; LR48877; P.L. Light
Systems, Beamsville, ON, Canada), 631-W LED fixtures (B20G50R45FR5; VYPR 2p; Fluence,
Austin, TX), 325-W LED fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville, CA),
600-W LED fixtures (B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a combination of
72-W LED fixtures (B120; HortiLED MULTI, P.L. Light Systems) and 625-W LED fixtures (R120;
LumiGrow Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro 650E; LumiGrow).
..................................................................................................................................................... 148
Table 4.4. Adjusted hue angle (h°), chroma (C), and Hunter CIELAB (L*, a*, b*) values at harvest
for godetia, snapdragon, and stock grown under six different supplemental lighting (SL)
treatments. SL treatments consisted of either 460-W HPS fixtures (HPS120; LR48877; P.L. Light
Systems, Beamsville, ON, Canada), 631-W LED fixtures (B20G50R45FR5; VYPR 2p; Fluence,
Austin, TX), 325-W LED fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville, CA),
600-W LED fixtures (B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a combination of
72-W LED fixtures (B120; HortiLED MULTI, P.L. Light Systems) and 625-W LED fixtures (R120;
LumiGrow Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro 650E; LumiGrow).
Letters indicate mean separations across treatments using Tukey-Kramer honestly significant
difference (HSD) test at P ≤0.05. .................................................................................................. 149
Table 4.5. Regression equations and R2 for time to visible bud, time to open flower, time to harvest,
and stem length at harvest of godetia 'Grace Rose Pink', stock 'Iron Rose', and snapdragon 'Potomac
Royal' in response to the estimated phytochrome photoequilibrium of each supplemental lighting
treatment. ** and *** indicate model significance at P <.001 and P <.0001, respectively. All
models are in the form of: ƒ = y0 + a*PPE + b*PPE2. ............................................................... 150
ix
LIST OF FIGURES
Figure 2.1. Effect of 9, 11, 12, 13, 14, 15, 16, 18, 24 h photoperiods or a 4-h night interruption
(NI) on the height (A; D), root dry mass (B; E), and shoot dry mass (C; F) of marigold ‘Xochi’
(Tagetes erecta) and witchgrass ‘Frosted Explosion’ (Panicum capillare) young plants. Black
symbols indicate means; error bars indicate standard error of the mean. NI means were excluded
from regressions. Figure 1-F presents data from replication 2 as trends from replication 1 were
not significant. Coefficients are presented in Table 2.7. ............................................................ 69
Figure 2.2. Effects of young-plant photoperiod (9, 12, 13, 14, 16, 18, 24 h, or a 4-h NI) and
finishing photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to visible flower bud of
witchgrass ‘Frosted Explosion’ (Panicum capillare). Figures represent (A) moderate-DLI-grown
(≈10 mol∙m–2∙d–1) cut flowers from replication 1, (B) very-low-DLI-grown (≈3 mol∙m–2∙d–1) cut
flowers from replication 1, (C) moderate-DLI-grown cut flowers from replication 2, and (D)
very-low-DLI-grown cut flowers from replication 2. Black symbols represent individual data
points for sequential photoperiods; red symbols represent means from NI treatments. NI means
were excluded from response surfaces. Model predictions are represented by response surfaces;
coefficients are presented in Table 2.8. .........................................................................................70
Figure 2.3. Effect of young-plant photoperiod (11, 13, 14, 15, 16, 24 h, or a 4-h NI) and/or finishing
photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to visible flower bud (TVB) of
marigold ‘Xochi’ (Tagetes erecta). Figures represent (A) the interaction between young-plant and
finishing photoperiod on TVB of plants from replication 1, (B) the effect of young-plant
photoperiod on TVB of plants from replication 2, and (C) the effect of finishing photoperiod on
TVB of plants from replication 2. In figure 3A, black symbols represent individual data points for
sequential photoperiods; red symbols represent means from NI treatments. NI means were
excluded from response surfaces and regressions. Coefficients are presented in Table 2.9. .......71
Figure 2.4. Effects of young-plant photoperiod [9, 12, 13, 14, 16, 18, 24 h, or a 4-h night
interruption (NI)] and finishing photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to
open flower of witchgrass ‘Frosted Explosion’ (Panicum capillare). Figures represent (A)
moderate-DLI-grown (≈10 mol∙m–2∙d–1) cut flowers from replication 1, (B) very-low-DLI-grown
(≈3 mol∙m–2∙d–1) cut flowers from replication 1, (C) moderate-DLI-grown cut flowers from
replication 2, and (D) very-low-DLI-grown cut flowers from replication 2. Black symbols
represent individual data points for sequential photoperiods; red symbols represent means from NI
treatments. NI means were excluded from response surfaces. Model predictions are represented by
response surfaces; coefficients are presented in Table 2.8. ...........................................................72
Figure 2.5. Effects of young-plant photoperiod [9, 12, 13, 14, 16, 18, 24 h, or a 4-h night
interruption (NI)] and finishing photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on stem
length of witchgrass ‘Frosted Explosion’ (Panicum capillare) at open flower. Figures represent
(A) moderate-DLI-grown (≈10 mol∙m–2∙d–1) cut flowers from replication 1, (B) very-low-DLI-
grown (≈3 mol∙m–2∙d–1) cut flowers from replication 1, (C) moderate-DLI-grown cut flowers from
replication 2, and (D) very-low-DLI-grown cut flowers from replication 2. Black symbols
x
represent individual data points for sequential photoperiods; red symbols represent means from NI
treatments. NI means were excluded from response surfaces. Model predictions are represented by
response surfaces; coefficients are presented in Table 2.8. ...........................................................73
Figure 3.1. Effect of 9-, 10-, 11-, 12-, 13-, 15-, and 16-h young-plant photoperiods on the height
(A; D), shoot dry mass (B; E), and root dry mass (C; F) of dianthus ‘Amazon Neon Cherry’
(Dianthus barbatus interspecific) young plant growth responses. Black symbols indicate means;
error bars indicate standard error of the mean. Coefficients are presented in Table 3.3. ........... 106
Figure 3.2. Effect of 9-, 10-, 11-, 12-, 13-, 15-, and 16-h young-plant photoperiods on the height
(A; D), shoot dry mass (B; E), and root dry mass (C; F) of dianthus ‘Amazon Rose Magic’
(Dianthus barbatus interspecific) young plant growth responses. Black symbols indicate means;
error bars indicate standard error of the mean. Coefficients are presented in Table 3.3. ........... 107
Figure 3.3. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to visible flower bud of dianthus
‘Amazon Neon Cherry’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-
grown cut flowers, (B) low-DLI-grown cut flowers from replication 1, and (C) low-DLI-grown
cut flowers from replication 2. Black symbols represent individual data points for sequential
photoperiods; red symbols represent means from NI treatments. NI means were excluded from
response surfaces. Model predictions are represented by response surfaces; coefficients are
presented in Table 3.4. ................................................................................................................ 108
Figure 3.4. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to visible flower bud of dianthus
‘Amazon Rose Magic’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-
grown cut flowers, (B) low-DLI-grown cut flowers from replication 1, and (C) low-DLI-grown
cut flowers from replication 2. Black symbols represent individual data points for sequential
photoperiods; red symbols represent means from NI treatments. NI means were excluded from
response surfaces. Model predictions are represented by response surfaces; coefficients are
presented in Table 3.5. ................................................................................................................ 109
Figure 3.5. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to open flower of dianthus ‘Amazon
Neon Cherry’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-grown cut
flowers, (B) low-DLI-grown cut flowers from replication 1, and (C) low-DLI-grown cut flowers
from replication 2. Black symbols represent individual data points for sequential photoperiods; red
symbols represent means from NI treatments. NI means were excluded from response surfaces.
Model predictions are represented by response surfaces; coefficients are presented in Table 3.4. ..
..................................................................................................................................................... 110
Figure 3.6. Individual effects of young-plant (9, 11, 12, 13, 15, or 16 h) or finishing photoperiod
(11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to open flower of dianthus ‘Amazon Rose Magic’
(Dianthus barbatus interspecific). Figures represent (A and B) moderate-DLI-grown cut flowers
and (C and D) low-DLI-grown cut flowers. Black symbols indicate means; error bars indicate
xi
standard error of the mean. NI means were excluded from regressions. Coefficients are presented
in Table 3.5. ................................................................................................................................. 111
Figure 3.7. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and/or finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to harvest of dianthus ‘Amazon Neon
Cherry’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-grown cut flowers
from replication 1, and (B) moderate-DLI-grown cut flowers from replication 2. Black symbols
represent individual data points for sequential photoperiods; red symbols represent means from NI
treatments. NI means were excluded from response surfaces. Model predictions are represented by
response surfaces; coefficients are presented in Tables 3.4. ....................................................... 112
Figure 3.8. Individual effects of young-plant (9, 11, 12, 13, 15, or 16 h) or finishing photoperiod
(11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to harvest of dianthus ‘Amazon Rose Magic’
(Dianthus barbatus interspecific) finished under a moderate DLI. Black symbols indicate means;
error bars indicate standard error of the mean. Coefficients are presented in Table 3.5. ..
..................................................................................................................................................... 113
Figure 3.9. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and/or finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on stem length at harvest of dianthus ‘Amazon
Neon Cherry’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-grown cut
flowers from replication 1, (B) low-DLI-grown cut flowers from replication 1 and 2, and (C)
moderate-DLI-grown cut flowers from replication 2. Black symbols in Fig. 9A and C represent
individual data points for sequential photoperiods; red symbols represent means from NI
treatments. Black symbols in Fig. 3.9B indicate means; error bars indicate standard error of the
mean. NI means were excluded from regressions. Model predictions are represented by response
surfaces or regressions; coefficients are presented in Table 3.4. ................................................ 114
Figure 3.10. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and/or finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on stem length at harvest of dianthus ‘Amazon
Rose Magic’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-grown cut
flowers, (B) low-DLI-grown cut flowers from replication 1, and (C) low-DLI-grown cut flowers
from replication 2. Black symbols in Fig. 3.10A indicate means; error bars indicate standard error
of the mean. Black symbols in Figs. 3.10B and C represent individual data points for sequential
photoperiods; red symbols represent means from NI treatments. NI means were excluded from
regressions. Model predictions are represented by response surfaces or regressions; coefficients
are presented in Table 3.5. .......................................................................................................... 115
Figure 3.11. Observational differences between dianthus ‘Amazon Rose Magic’ (Dianthus
barbatus interspecific) cut flowers finished under a moderate DLI or a low DLI. ..................... 116
Figure 4.1. Emission spectra of supplemental lighting (SL) fixtures utilized throughout the study.
SL treatments consisted of either 460-W HPS fixtures (HPS120; LR48877; P.L. Light Systems,
Beamsville, ON, Canada), 631-W LED fixtures (B20G50R45FR5; VYPR 2p; Fluence, Austin, TX),
325-W LED fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville, CA), 600-W LED
fixtures (B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a combination of 72-W LED
xii
fixtures (B120; HortiLED MULTI, P.L. Light Systems) and 625-W LED fixtures (R120; LumiGrow
Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro 650E; LumiGrow). .............. 151
Figure 4.2. Time to visible flower bud, time to open flower, time to harvest, and stem length at
harvest of godetia ‘Grace Rose Pink’ and stock ‘Iron Rose’ in response to SL spectrum, pooled
over two replications. SL treatments consisted of either 460-W HPS fixtures (HPS120; LR48877;
P.L. Light Systems, Beamsville, ON, Canada), 631-W LED fixtures (B20G50R45FR5; VYPR 2p;
Fluence, Austin, TX), 325-W LED fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow,
Emeryville, CA), 600-W LED fixtures (B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a
combination of 72-W LED fixtures (B120; HortiLED MULTI, P.L. Light Systems) and 625-W LED
fixtures (R120; LumiGrow Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro 650E;
LumiGrow).Letters indicate mean separations across treatments using Tukey-Kramer honestly
significant difference (HSD) test at P ≤0.05. Bars represent the mean and error bars indicate
standard error. .............................................................................................................................. 152
Figure 4.3. Time to visible flower bud, time to open flower, time to harvest, and stem length at
harvest of snapdragon ‘Potomac Royal’ in response to SL spectrum over two replications. SL
treatments consisted of either 460-W HPS fixtures (HPS120; LR48877; P.L. Light Systems,
Beamsville, ON, Canada), 631-W LED fixtures (B20G50R45FR5; VYPR 2p; Fluence, Austin, TX),
325-W LED fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville, CA), 600-W
LED fixtures (B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a combination of 72-W
LED fixtures (B120; HortiLED MULTI, P.L. Light Systems) and 625-W LED fixtures (R120;
LumiGrow Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro 650E; LumiGrow).
Letters indicate mean separations across treatments using Tukey-Kramer honestly significant
difference (HSD) test at P ≤0.05. Bars represent the mean and error bars indicate standard error.
..................................................................................................................................................... 153
Figure 4.4. Influence of estimated phytochrome photoequilibrium (PFR/PR+PFR) of supplemental
lighting treatments on time to visible bud, time to open flower, time to harvest, and stem length at
harvest of godetia 'Grace Rose Pink' and stock 'Iron Rose'. Black symbols represent means; error
bars represent standard error. R2 values are presented; ** and *** indicate model significance at P
<.001 and P <.0001, respectively. Coefficients are presented in Table 4.5. .............................. 154
Figure 4.5. Influence of estimated phytochrome photoequilibrium (PFR/PR+PFR) of supplemental
lighting treatments on time to visible bud, time to open flower, time to harvest, and stem length at
harvest of snapdragon 'Potomac Royal' during replications 1 and 2. Black symbols represent
means; error bars represent standard error. R2 values are presented; ** and *** indicate model
significance at P <.001 and P <.0001, respectively. Coefficients are presented in Table 4.5. . .. 155
xiii
SECTION 1
1
Literature Review: Manipulating photon flux density, photon spectrum, and photoperiod to
improve the greenhouse production of specialty cut flowers
Introduction
Cut flowers are economically important crops that carry an emotional sentiment and
symbolic meaning (Faust and Dole, 2021; Yue and Behe, 2010; Yue and Hall, 2010; Armitage
and Laushman, 2003). Mother’s Day and Valentine’s Day accounted for 24% and 20%,
respectively, of annual cut flower transactions in 2015 (PMAFMI, 2016). Not only are cut
flowers popular for calendar occasions and holidays, but they are purchased for birthdays, home
decoration, memorial occasions, and other occasions independent of calendar date (Yue and
Behe, 2010; Yue and Hall, 2010). Cut flowers are generally distinguished between two groups,
traditional and specialty. Traditional cut flowers include roses (Rosa spp.), chrysanthemums
(Chrysanthemum ×morifolium), and carnations (Dianthus caryophyllus), which have historically
accounted for the largest portion of international cut flower production (Armitage and
Laushman, 2003; Yue and Hall, 2010). The Association of Specialty Cut Flower Growers
defines specialty cut flowers as those that are not on the market on a regular basis, or only for an
exceptionally short time period (Armitage and Laushman, 2003), such as celosia (Celosia spp.),
larkspur (Delphinium spp.), and zinnia (Zinnia spp.; Sparks, 2020). Furthermore, many specialty
cut flower varieties are less resilient than traditional cut flowers and they often do not ship well
as they have a relatively short postharvest life (Owen et al., 2016; Wien, 2009; Redman et al.,
2002).
From 2015 to 2018, the number of domestic cut flower producers with annual sales
≥$100,000 increased from 444 to 532, respectively, with a wholesale value of $374 million in
2018 (USDA, 2019). California accounted for $288 million (77%) of domestic production due to
the coastal climates of its central and south coast counties (Carman, 2007). However, the
2
importation of traditional cut flowers from Colombia and Ecuador to the United States (U.S.) has
significantly reduced domestic production volume over time (Loyola et al., 2019b; PMAFMI,
2016; USITC, 2003). Due to their proximity to the equator and high altitude, these countries
have consistently high radiation intensities and moderate temperatures, favorable to the growth
of many cut flower varieties (Faust and Dole, 2021). Additionally, these countries have
consistent photoperiods of ≈12 h, which is favorable as long day (LD) or short day (SD)
conditions can easily be provided to either delay or induce flowering in various cut flower genera
(Faust, 2021, personal communication; Faust and Dole, 2021). These climactic conditions, as
well as inexpensive labor relative to North America, minimal greenhouse structures, and low
energy costs, reduce overall production costs (Faust and Dole, 2021; Bergmann, 2019; Kelly,
1991). As a result, wholesale prices are significantly lower than they are for most domestically
produced traditional cut flowers. This price difference has made it difficult for domestic growers
to compete by growing solely traditional cut flowers. In fact, traditional cut flowers only
accounted for 7% of domestic cut flower production in 2018 (USDA, 2019).
With approximately 68% and 42% of floral sales being conducted in person or being
characterized as impulsive, respectively, the aesthetic quality of cut flowers is of paramount
importance (PMAFMI, 2016). Although wholesale standards for cut flower quality vary among
countries, organizations, and producers, high-quality cut flowers are typically have thick stems
and large flowers (Armitage and Laushman, 2003). In 1996, the Floral Marketing Association
(FMA) and Society of American Florists (SAF) graded cut flowers through several parameters
including minimum stem length, stem strength, stem deviation, flower diameter, and flower
count per stem (FMA/SAF, 1996). Although specifications vary between cut flower varieties,
longer stem length and strength, consistent stem straightness, large flower size, and high flower
3
count are typically desirable. Furthermore, high-quality cut flowers are free of pests, diseases,
and blemishes, have a long vase life, and haven’t been exposed to ethylene (FMA/SAF, 1996).
While minimum desirable stem length varies between cut flower varieties, an average minimum
stem length of 30 cm for cut flowers sold at farmers markets has been utilized for recent cut
flower production research to assess stem marketability (Ortiz et. al, 2012; Owen et al., 2016).
However, Ahmad et al. (2017) indicated that stem lengths ≥60 cm are preferred by florists and
can be sold for higher prices.
Imported specialty cut flowers may be of lesser quality than those produced domestically,
as their quality decreases during shipping (Kaiser and Ernst, 2018). This is undesirable, as the
top demands from mass market floral buyers in 2015 included consistent cut flower quality
throughout the year, and a selection of the freshest cut flowers available. Moreover, consumers
sought freshness and diversity in their selection of cut flowers (PMAFMI, 2016). In a separate
survey, 67% of cut flower producers in North Carolina experienced an increase in demand for
locally produced fresh cut flowers (Granitz, 2014). Short et al. (2017) also indicated that
consumers favor cut flowers grown within 100 miles of the point of sale. Additionally, a 2021
survey on American cut flower preference reported that 57% of respondents found it important
that their flowers were produced domestically. Furthermore, 58% of respondents preferred
locally-grown cut flowers (Prinzing, 2021).
Domestic specialty cut flower production is a promising method of providing consumers
with fresher, more diverse and exciting product while affording local growers a market that
international producers cannot compete with. Growers who have begun to capitalize on these
trends have prospered, as high-quality specialty cut flowers can reach a value of $25,000 to
$30,000 per acre (Byczynski, 2008). However, not all regions of the U.S. permit year-round
4
production of specialty cut flowers. While coastal climates such as those in California permit
domestic year-round field production of cut flowers (Kelly, 1991), regions with temperate
climates and reduced solar radiation during the winter, such as the Midwestern and Northern
U.S., do not permit field or high tunnel cut flower production year-round, although the desire for
fresh and local flowers persists (Ortiz et al., 2012; Owen et al., 2016). Therefore, by investing in
protected structures such as controlled-environment greenhouses, local growers in these
temperate climates can produce specialty cut flowers for local markets year-round.
High Tunnel Cut Flower Production
One option available to growers seeking protected cultivation for their cut flower crop is
high tunnels, which are significantly less expensive to install and operate than traditional
greenhouses. High tunnels are semi-permanent structures that are traditionally composed of a
metal pipe framework and glazed with four to six-mm thick polyethylene film. These structures
are not equipped with automated heating and ventilation systems, or other mechanized systems
such as automated irrigation, and supplemental and photoperiodic lighting systems (Lamont,
2009; Lopez, 2016; Ortiz et al., 2012; Owen et al., 2016).
High tunnels are valuable structures for season extension production, allowing for earlier
harvests in the spring and later harvests in the fall compared to field production in temperate
regions (Lamont, 2009; Owen et al., 2016; Wien, 2009). They can also reduce the impact of
inclement weather on a crop; polyethylene glazing protects plants from precipitation and strong
winds that have the potential to significantly reduce crop quality (Owen et al., 2016; Wien,
2009). Additionally, high tunnels can yield a greater quantity of stems with quality superior to
field-grown cut flowers in terms of stem length and stem caliper, although these responses are
5
crop specific. For example, the number of marketable stems per m2 of high-tunnel-grown
snapdragon ‘Rocket Red’ (Antirrhinum majus), dianthus ‘Amazon Neon Cherry’ (Dianthus
barabatus interspecific) and zinnia ‘Benary’s Giant Scarlet’ (Zinnia elegans) over the course of a
study were 26, 185, and 192 stems greater, respectively, compared to field production yield.
Additionally, stem length of high-tunnel-grown snapdragon and zinnia were 33% and 32%
longer, respectively, compared to field-grown plants. A similar trend was found for inflorescence
length; high-tunnel grown snapdragon and zinnia had inflorescences 28% and 5% longer,
respectively, compared to field-grown plants (Ortiz et al., 2012).
Similarly, Owen et al. (2016) reported that the stem lengths of high tunnel grown bells of
Ireland (Moluccella laevis), celosia ‘Bombay Firosa’ (Celosia cristata), dianthus ‘Amazon Neon
Purple’, gomphrena ‘Fireworks’ (Gomphrena pulchella), and matricaria ‘Vegmo Snowball
Extra’ (Tanacetum parthenium) were 15%, 16%, 18%, 44%, and 19% taller, respectively,
compared to field-grown stems. However, stem length was not influenced by production system
for bellflower ‘Campana Deep Blue’ (Campanula carpatica) or snapdragon ‘Potomac
Lavender’. The authors hypothesized that the stem elongation of several of the high tunnel
grown cut flowers could have been a result of reduced air movement in the high tunnel
environment when compared to the field (Ortiz et al., 2012; Owen et al., 2016).
Another feasible explanation of increased stem length under high tunnels is related to the
total radiation intensity received by the crops. Single-layer polyethylene film utilized to glaze
high tunnels initially transmits ≈85% of intercepted radiation to the crops inside (Both and Faust,
2017) when compared to field-grown crops, which can result in a taller stem length due in part to
reduced radiation intensities (Dole and Warner, 2017). This explanation is supported by
Armitage (1991), who investigated the influence of several grades of neutral shading on the
6
finished stem length of field-grown American basketflower ‘Jolly Joker’ (Centaurea americana),
globe thistle ‘Taplow Blue’ (Echinops ritro), sea holly (Eryngium planum), and four varieties of
calla lily (Zantedeschia hybrida). Neutral shading decreased the photosynthetic photon flux
density (PPFD) within each experimental plot without changing the spectral quality of the
transmitted radiation. Average stem length was 23%, 22%, 9%, and 54% taller, respectively, for
American basketflower ‘Jolly Joker,’ globe thistle ‘Taplow Blue,’ sea holly, and four varieties of
calla lily grown under 67% shade in comparison to those grown without shade (Armitage, 1991).
Floriculture crop production in high tunnels can have drawbacks. For instance, high
tunnels typically lack an electricity source for supplemental lighting (SL) or photoperiodic
lighting, used to regulate the growth and development of crops. As a result, high tunnel cut
flower growers are limited to growing crops that will flower under the natural daylength of the
given season. Additionally, although photo-selective plastics and filters exist (Runkle and Heins,
2001), many high tunnel growers do not utilize them, and thus cannot manipulate the radiation
quality received by their crops.
In northern latitudes, crops grown in high tunnels can experience both extremely high and
low air temperatures, depending on the time of year, due to the lack of temperature regulating
equipment (Wien, 2009). This can lead to reduced crop quality and in some cases crop death,
unless ventilation is practiced via roll-up sides during the spring and summer months, and heat is
supplied during the winter months. Furthermore, because temperature cannot be easily managed
within high tunnels, growers cannot reliably manipulate it to influence crop growth and
development, which can be risky when scheduling to have flowers ready for sale on pre-
determined dates. The lack of temperature, humidity, and air circulation regulating technology
can also lead to disease proliferation (USDA NRCS, 2014). Lastly, if high tunnel cut flower
7
growers produce their crops in poorly managed soil, fertilizer salt accumulation can occur, which
may lead to poor seed germination and the inhibition of plant growth (Knewtson et al., 2010).
The utilization of controlled-environment greenhouses for cut flower production provides
solutions for these problems. The ability to manipulate temperature; and radiation quality,
intensity, and duration in a greenhouse gives growers the potential to produce high-quality cut
flowers for local markets year round.
Controlled-Environment Greenhouse Cut Flower Production
Photoperiodic Crop Lighting in Controlled Environment Greenhouses
Photoperiodism is broadly defined as responses to the length of periods of light and
darkness that enable living organisms to adapt to seasonal changes in their environment (Thomas
and Vince-Prue, 1997). For instance, photoperiodic flowering responses evolved so angiosperms
could synchronize flowering with pollinator activity, increasing the chance of pollination and
subsequent seeding (Thomas, 2017). The photoperiod is the number of consecutive hours of
radiation in a 24-h period (Dole and Wilkins, 2005; Runkle, 2008b). Plants are categorized as
short-day (SDPs), long-day (LDPs), or day-neutral plants (DNPs), depending on their flowering
response to night length (Craig and Runkle, 2012; Erwin and Warner, 2002; Mattson and Erwin,
2004). Although plants biologically respond to night length, photoperiodic requirements are
conventionally quantified by a plant’s critical photoperiod. Horticulturally, SDPs and LDPs can
be classified as having an obligate (qualitative) photoperiodic response, where a genotype-
specific critical photoperiod is required to induce flowering, or a facultative (quantitative)
photoperiodic response, where a critical photoperiod is not required, but accelerates the rate of
flower development (Craig and Runkle, 2012; Mattson and Erwin, 2004). DNPs flower
8
regardless of photoperiod if all other environmental factors are favorable (Craig and Runkle,
2012; Mattson and Erwin, 2004).
Photoperiodic responses are mediated in part by photoreceptors such as phytochromes
(Borthwick and Hendricks, 1960; Craig and Runkle, 2012; Mer and Attri, 2015; Piringer and
Borthwick, 1961). Phytochromes exist within the cytosol of plant cells and are composed of
proteins bound to radiation-absorbing chromophores called phytochromobilin (Hernández and
Kubota, 2017; Rüdiger and Thümmler, 1994). Phytochromes in light-grown plants exist in two
interconvertible conformations; a red radiation (R) absorbing form (PR; 600 to 700 nm, peak
absorption at 660 nm) and a far-red radiation (FR) absorbing form (PFR; 700 to 800 nm, peak
absorption at 735 nm; Craig and Runkle, 2012, 2013; Sager et al., 1988). PR will convert to PFR
upon absorption of R radiation. Conversely, PFR will convert to PR upon absorption of FR
radiation or upon degradation during the dark period (Nelson, 1991; Sager et al, 1988; Walters et
al., 2019). PR and PFR are typically denoted as the “inactive” and “active” conformations of
phytochrome, respectively, as PFR triggers physiological responses within plant cell nuclei (Craig
and Runkle, 2012; Hernández and Kubota, 2017; Nelson, 1991; Runkle and Heins, 2001).
Thus, the duration of the photoperiod and skotoperiod, as well as radiation quality during
the photoperiod, influence the ratio of phytochrome conformations, known as the phytochrome
photoequilibrium (PPE = PFR/PR+FR; Craig and Runkle, 2012, 2013; Hernández and Kubota,
2017; Sager et al., 1988). This ratio subsequently influences flowering in at least some
photoperiodic plants, among other phytochrome-regulated photomorphogenic responses (Craig
and Runkle, 2012, 2013; Hernández and Kubota, 2017; Runkle and Heins, 2006). Natural SDs
and uninterrupted long nights result in a relatively low PPE, generally promoting flowering in
SDPs and inhibiting flowering in LDPs (Craig and Runkle, 2013). Conversely, LDs and short
9
nights result in a relatively high PPE, generally conducive with flowering in LDPs and the
inhibition of flowering in SDPs (Craig and Runkle, 2013). The PPEs required for the transition
from vegetative to reproductive growth vary based on genotype and are related to a crop's critical
photoperiod.
A 2019 survey reported that crop timing, including controlling time to flower, was the
second most important production issue that North American specialty cut flower growers face
after insect pest management (Loyola et al., 2019a). This report indicates that growers could
benefit from unbiased research regarding the manipulation of the light environment and how it
influences the growth and development of specialty cut flowers. Greenhouse technologies exist
that can provide specific photoperiods to induce flowering responses in various crops year round
(Hammer, 2011; Moe et al., 2006; Sherrard, 2011). The utilization of black-out curtains to
provide SD conditions and photoperiodic lighting systems to provide LD conditions has proved
useful in truncating or extending the natural photoperiod, respectively, in greenhouses in regions
and seasons where the natural photoperiod is not conducive with flowering (Runkle et al., 2017).
It is important to have the ability to establish LD or SD conditions when growing LDPs
or SDPs during seasons with natural SDs or LDs, respectively. Fortunately, a myriad of LD
lighting techniques can be used strategically in controlled-environment greenhouses to either
promote flowering of LDPs or inhibit flowering of SDPs. Opaque black-out curtains can also be
utilized in greenhouses to promote flowering of SDPs or inhibit flowering of LDPs regardless of
season. LDs can be simulated in the greenhouse by utilizing cyclic lighting (CL), night-
interruption (NI) lighting, or day-extension (DE) lighting (Runkle and Both, 2017). While DE
lighting is utilized at the end of the day to effectively extend the natural photoperiod, NI lighting
and CL operate for several hours during the middle of the dark period at low (1 to 2
10
µmol∙m– 2∙s- 1) or high radiation intensities, respectively, simulating LDs by moderately
increasing the PPE and saturating the flowering response (Craig and Runkle, 2016; Owen et. al,
2018; Runkle and Both, 2017; Whitman et. al, 1998).
These low and high intensity PL techniques are most effective when lamps emit both R
and FR radiation, as FR-deficient lamps can delay or inhibit flowering in some LDPs (Craig and
Runkle, 2016; Craig and Runkle, 2012; Runkle and Heins, 2006; Walters et al., 2019). This is
because flowering in LDPs is theorized to be regulated by PFR in one of two ways (Craig and
Runkle, 2016; Thomas and Vince-Prue, 1997). Some LDPs flower once a certain threshold of
PFR, or greater, is reached, from light sources with moderate to high R:FR ratios. Conversely,
excessive concentrations of PFR can inhibit flowering in other LDPs. Craig and Runkle (2016)
concluded that an intermediate PPE, provided by a light source providing both R and FR
radiation, could elicit LD responses in both types of LDPs, as the concentrations of PFR within
plant cells would be high enough to initiate flowering in all LDPs, but not high enough for
flower inhibition to occur for LDPs in the latter designation. Thus, the resulting PPE in plants
under these conditions would moderately lower in comparison to that elicited by FR-deficient
lamps (Craig and Runkle, 2016; Runkle and Heins, 2001). However, it is typically not as low as
the PPE characteristic of short photoperiods and long, uninterrupted dark periods.
Blanchard and Runkle (2009) demonstrated that cyclic lighting provided by an oscillating
high-pressure sodium (HPS) lamp emitting a radiation intensity of ≈2.4 µmol∙m–2∙s–1 inhibited
flowering of the facultative SDPs chrysanthemum ‘Auburn’ and ‘Bianca’ (Chrysanthemum
×grandiflorum). The HPS lamp operated continuously during a 4-h NI period (2230 to 0230 HR),
with crops grown 1, 4, 7, 10, and 13 m from the radiation source. Under a 9-h SD,
chrysanthemum ‘Auburn’ flowered 30 and 15 d faster than crops grown 10 and 13 m,
11
respectively, from the HPS lamp, and chrysanthemum ‘Bianca’ flowered 30, 13, and 7 d faster
than crops grown 7, 10, and 14 m, respectively, from the HPS lamp. Chrysanthemum ‘Auburn’
did not flower when grown 1, 4, and 7 m from the HPS lamp, and chrysanthemum ‘Bianca’ did
not flower when grown 1 and 4 m from the HPS lamp (Blanchard and Runkle, 2009).
Some photoperiod studies have been conducted on genera grown as cut flowers. For
example, a 24-h photoperiod during the winter promoted 100% flowering of the facultative LDP
gladiolus ‘Kundard White’ (Gladiolus spp.) with an average of 9 florets per spike in comparison
to 7% flowering and an average of 2 florets per spike for crops grown under an 8-h SD (Kosugi,
1962). In a separate study, 65% of gladiolus ‘Spotlight’ grown under 2-h of NI lighting flowered
compared to no plants under an 8-h SD (Kosugi, 1962). Additionally, floral initiation of
anemone ‘de Caen’ (Anemone coronaria) under a 4-h NI from 2200 to 0200 HR was hastened by
10 d compared to plants under a natural day length ranging from 11 to 12 h (Ohkawa, 1987).
In another study, Blacquière et al. (2002) demonstrated that matricaria ‘Snowball’ and
‘Single Vegmo’ exhibited an obligate LDP response, as flower induction occurred under 7- to 8-
h DE lighting creating an 18-h photoperiod. Additionally, NI lighting applied at an intensity of 2
µmol∙m–2∙s–1 at various points throughout the dark period promoted floral initiation. For instance,
flowering of both matricaria cultivars was promoted after ≈50 d when a 2-h NI was initiated 8 h
after the beginning of a dark period. However, flowering was delayed by ≈6 d under a 14-h
photoperiod created with 2-h of DE lighting beginning before dawn in comparison to the NI
beginning 8 h after the beginning of the dark period. Lastly, 2 h of DE lighting beginning at dusk
did not promote flowering in either cultivar (Blacquière et al., 2002). Furthermore, the SDP blue
spiraea (Caryopteris incana) grown as a cut flower reached anthesis 11 d earlier on average
when grown under an 8-h photoperiod in comparison to a 12-h photoperiod (Armitage and Son,
12
1992). Additionally, an 8-h photoperiod hastened flowering of the SDPs marigold ‘Discovery
Orange’ and ‘Discovery Yellow’ (Tagetes erecta) by 13 and 5 d, respectively, compared to
plants grown under a 14-h photoperiod (Köksal et al., 2017).
Photoperiod control isn’t only useful for promoting flowering in photoperiodic plants
when the natural daylength isn’t conducive with flowering. SDPs produced in high tunnels early
in the growing season may be exposed to naturally SD lengths, resulting in premature flowering
and a decrease in finished quality (Warner, 2009), including a final stem length shorter than the
minimum marketable standard (Ortiz et al., 2012; Owen et al., 2016).
When a plant overcomes juvenility and photoperiodic flowering requirements are met
prematurely, it is not always possible to revert plants to their vegetative state with just a simple
change in the lighting strategy to non-inductive photoperiods (Runkle, 2008a; Warner, 2009;
Runkle, 2018). In fact, cultural techniques such as limited-inductive photoperiod have been
developed, which are based on this principle (Damann and Lyons, 1993; Runkle, 2008a; Warner,
2009). However, some obligate photoperiodic crops, such as some cultivars of pansy and petunia
(LDPs) and butterfly weed (Asclepias spp.; SDP) will cease floral development when moved to
non-inductive photoperiods (Runkle, 2008a), although this effect has rarely been seen with other
plants (Runkle, 2021, personal communication). For plants that have initiated flower buds
prematurely, several applications of flower-inhibiting plant-growth regulators such as ethephon
over time can be successful at keeping flowers at bay until flowering is desired, although this
effect varies by crop (Styer, 2002). Moreover, ethephon applications are generally undesirable
for cut flower cultivation as ethephon can reduce stem lengths by suppressing apical dominance
(Runkle, 2013). Manipulating radiation duration to prevent early flowering as soon as the
juvenile period concludes, or immediately after germination if the juvenile period is unknown,
13
can help growers avoid unnecessary plant-growth regulator applications down the line, not only
reducing overall chemical usage, but preserving crop quality and reducing labor. Therefore, it is
generally recommended that cut flowers receive a non-inductive photoperiod until stem length is
sufficient and flowering is desired (Dole and Warner, 2017; Porat et al., 1995).
For instance, when plugs of the SDP celosia Kurume Series ‘Sakata Pride’ (Celosia
cristata) were grown under incandescent lamps to create a 24-h photoperiod, subsequent flower
primordia initiation was delayed up to 15 d compared to those grown under a 15-h photoperiod
during the seedling stage (Goto and Muraoka, 2008). In a separate study, Warner (2009) reported
that celosia ‘Gloria Scarlet’ (Celosia plumosa) grown under a 4-h NI (2200 to 0200 HR) had ≈10
more nodes beneath the terminal inflorescence and ≈5 more lateral inflorescences when
compared to crops grown under a 9-h SD. Additionally, celosia ‘Gloria Scarlet’ grown under NI
lighting had a shoot dry mass ≈4.8 g greater than crops grown under continuous SDs. Although
flowering was hastened under SDs, finished plant quality was higher under NI lighting (Warner,
2009). In a separate study, pinched celosia ‘Cultivar’ (Celosia plusoma) grown under a 16-h
photoperiod for 3 weeks, and then a 8-h photoperiod for 29 d were ≈183% taller than plants
grown under continuous short days for 50 d (Porat et al., 1995).
Extending the photoperiod during the beginning of a SDP crop cycle can delay flowering,
ensuring that finished stems of are sufficient length and caliper. Issues with premature flowering
may be encountered when cultivating LDPs later in the growing season when natural day lengths
become longer, although truncating the day length to photoperiods non-conducive to flowering
can prevent premature flowering, rather than the aforementioned techniques.
14
Radiation Intensity in Controlled Environment Greenhouses
Radiation is the driving force of photosynthesis, and an increase in radiation intensity
generally promotes a greater accumulation of plant biomass until a crop-specific radiation
saturation point is reached (Faust, 2011). Photosynthetically active radiation (PAR) has
traditionally been defined as photons with wavelengths between 400 and 700 nm. However,
recent research has demonstrated that although FR radiation is a poor photosynthetic stimulus
when applied alone (Zhen et al., 2021), it can contribute directly and indirectly to photosynthesis
by expanding leaf area and working in synergy with photons within the traditional range of PAR
(Zhen and Bugbee, 2020a, b; Zhen and van Iersel, 2016). This phenomenon was demonstrated
across a diverse range of genera, leading to the advent of the term “ePAR”, referring to an
extension of PAR from 400 to 750 nm (Pazuki et al., 2017; Zhen and Bugbee, 2020a). Although
the implications of these findings are promising, this phenomenon is yet to be widely adopted
and utilized in the scientific community.
The total PAR received over a particular area in a day can be quantified as the
photosynthetic daily light integral (DLI). The DLI is analogous to a rain gauge– where a rain
gauge measures the total amount of precipitation that is recorded in a set period, the DLI
accounts for the summation of PAR that an area of one square meter receives over a 24-h period
(Torres and Lopez, 2012). In addition to an increase in biomass; improved lateral branching,
flower quantity, flower size, and rate of flowering is typically observed as the DLI increases
(Faust et al., 2005; Lopez, 2018; Pramuk and Runkle, 2005). Additionally, a low DLI can
promote stem elongation in some cut flower cultivars (Armitage, 1991; Treder, 2003; Treder and
Kubik, 2000).
15
The outdoor solar DLI between 35° and 45° N latitude can drop to as low as 5 to 10
mol∙m–2∙d–1 during the winter months (Korczynski et al., 2002). Furthermore, the DLI within
protected cultivation structures can be reduced by 40% or more because of reflection from
glazing, as well as shading from supporting infrastructure (Lopez and Runkle, 2008). Thus, the
DLI within a protected cultivation structure can be reduced to ≤5 mol∙m–2∙d–1 during the winter,
and even lower during periods of prolonged cloudy weather (Lopez and Runkle, 2008). This
presents a challenge to high tunnel and greenhouse cut flower growers that begin their extended
growing season by starting seedlings during this time period (Lopez and Olberg, 2016).
SL can be utilized in greenhouses to provide adequate DLIs for young-plant and finished
production for a variety of cut flower crops year-round. It is generally recommended that most
floriculture crops receive a DLI of 4 to 6 mol∙m–2∙d–1 with a PPFD of 100 to 200 µmol∙m–2∙s–1
under a 12-h photoperiod during the preliminary stages of propagation and root development
(Torres and Lopez, 2011). Adequate DLIs during young-plant production are critical for seedling
biomass accumulation as well as finished plant quality, as low radiation intensities during early
developmental stages of a plant’s life can decrease the quality of finished plants (Pramuk and
Runkle, 2005; Shillo and Halevy, 1976; Torres and Lopez, 2011). For instance, Shillo and
Halevy (1976) reported that a 5-d exposure of 80% shade on young gladiolus ‘Dr. Fleming’
corms that had 1 to 2 leaves resulted in 57% flowering during finishing in comparison to corms
that received no shade, in which 80% of plants flowered during finishing.
For most crops, moderate DLIs of 8 to 12 mol∙m–2∙d–1 during propagation can lead to
higher seedling quality and faster crop turnover throughout the season (Pramuk and Runkle,
2005; Oh et al., 2010). However, high-light (shade-avoiding) crops may require relatively higher
DLIs for acceptable plant quality. For example, stem diameter of yellow trumpet bush ‘Mayan
16
Gold’ (Tecoma stans) seedlings increased by 133% when the DLI increased from 0.8 to 25.2
mol∙m–2∙d–1 (Torres and Lopez, 2011). Additionally, relative leaf chlorophyll content increased
by 34% as the DLI increased from 0.8 to 15.6 mol∙m–2∙d–1 and root dry mass increased twenty-
four-fold as the DLI increased from 0.8 to 25.2 mol∙m–2∙d–1 (Torres and Lopez, 2011). Similarly,
Pramuk and Runkle (2005) reported that celosia ‘Gloria Mix’ (Celosia argentea var. plumosa),
and marigold ‘Bonanza Yellow’ (T. patula) flowered 10 and 5 d earlier, respectively, when
seedlings were grown under a DLI of 14.2 mol∙m–2∙d–1 compared to those grown under a DLI of
4.1 mol∙m–2∙d–1. In addition, celosia ‘Gloria Mix’ seedling height decreased by ≈4 cm as the DLI
increased from 4.1 to 14.2 mol∙m–2∙d–1 (Pramuk and Runkle, 2005).
Utilizing SL during the finishing stage can benefit the finished quality of many
floriculture crops. Begonia ‘Vodka Cocktail’ (Begonia ×semperflorens-cultorum), impatiens
‘Cajun Red’ (Impatiens walleriana), petunia ‘Apple Blossom’ (Petunia ×hybrida), vinca ‘Pacific
Lilac’ (Catharanthus roseus), and zinnia ‘Dreamland Rose’ produced 9.4, 3.4, 3.0, 2.2, and 0.5
more flowers on average as the DLI increased from 5 to 19 mol∙m–2∙d–1 (Faust et al., 2005).
Marigold ‘Antigua American Orange’ produced 1.6 more flowers on average as the DLI
increased from 19 to 43 mol∙m–2∙d–1. Furthermore, total dry mass increased as the DLI increased
from 5 to 43 mol∙m–2∙d–1 for all species (Faust et al., 2005).
Similar effects can be found when manipulating radiation quantity during finished cut
flower culture. For instance, time to flower of oriental lily ‘Laura Lee’ (Lilium spp.) decreased
by an average of 21 d and plants were 19.7% shorter when 60 µmol∙m–2∙s–1 of SL was provided
for 5 h∙d–1 with HPS lamps, compared to plants grown under natural radiation intensities (Treder,
2003). Treder and Kubik (2000) observed that in comparison to crops grown under natural
radiation intensities, the fresh weight, length of the first open flower, and total length of buds per
17
stem of finished oriental lily ‘Star Gazer’ (Lilium spp.) grown under 60 µmol∙m–2∙s–1 of SL
provided by HPS lamps for 10 h∙d–1 increased by 29%, 17%, and 51%, respectively. Moreover,
the average fresh mass of crops exposed to SL was 17 g higher than the crops grown under
natural radiation intensities, and time to flower was reduced by 22 d. Crops grown under SL
were 8% shorter, on average, than those grown without (Treder and Kubik, 2000). Although
height was slightly reduced under SL for crops in both of the previously described studies, it
remained well above the marketable standard for stem length.
Autio (2000) examined the effects of SL on gerbera ‘Estelle’ and ‘Ximena’ (Gerbera
×cantabrigensis) and reported that yield increased by 13 and 10 stems, respectively, as SL
increased the natural DLI by 3.2 to 6.5 mol∙m–2∙d–1 during a 12-h photoperiod. Similarly, total
dry mass per plant increased by 89% and 56% for gerbera ‘Estelle’ and ‘Ximena’ respectively, as
SL increased the natural DLI by 3.2 to 6.5 mol∙m–2∙d–1 (Autio, 2000). In a separate study,
Llewellyn et al. (2020) investigated the influence of SL during the winter on the maturation time
of greenhouse-grown gerbera ‘Ultima’ (Gerbera jamesonii) and yield and finished cut flower
quality of gerbera ‘Panama’. The maturation time (time between flower bud initiation and
harvest) of gerbera ‘Ultima’ decreased by 4 d as the total DLI increased from 5.3 to 11.3
mol∙m– 2∙d–1. Gerbera ‘Panama’ stem length was 9% shorter and flower diameter increased by
11% as the total DLI increased from 5.3 to 11.3 mol∙m–2∙d–1. Additionally, stem yield of gerbera
‘Panama’ increased by 40% as the total DLI increased from 5.3 to 11.3 mol∙m–2∙d–1 (Llewellyn et
al., 2020).
Lugasi-Ben-Hamo et al. (2010) reported that an increase in the DLI resulted in a higher
yield per m2 of greenhouse-grown lisianthus ‘Echo Champaign’ and ‘Rosita White’ (Eustoma
spp.). Lisianthus ‘Echo Champagne’ produced 12 more stems per m2 with an average of 1 more
18
bud per stem when grown under 67% shade for 5 weeks, compared to plants grown under 88%
shade for 5 weeks. Similarly, lisianthus ‘Rosita White’ produced 2 more stems per m2 with an
average of 3 more buds per stem when grown under 67% shade for 5 weeks, when compared to
plants grown under 88% shade for 5 weeks (Lugasi-Ben-Hamo et al., 2010).
Radiation Quality in Controlled-Environment Greenhouses
Growers who utilize SL can select from a broad range of commercially available fixtures.
However, different fixtures can influence crop growth and development in very different ways; a
function of the variation in their radiation spectra. Thus, it is important that growers understand
the impact that radiation quality can have on their crops when deciding which SL fixtures to
invest in.
White (W) radiation is composed of various colors, each correlating with a specific
waveband on the electromagnetic spectrum (Faust, 2011; Overheim and Wagner, 1982). Of these
wavebands, those included in PAR are utilized by plants for photosynthetic processes. While
PAR is integral for photosynthetic processes, it only accounts for 43% of solar radiation (Lopez
et al., 2017). Not every waveband from a radiation source is visible to the human eye or utilized
by plants for photosynthesis. However, the spectrum emitted by a radiation source can
significantly influence plant growth and development, with some wavelengths acting as a
morphological or developmental signal for plants rather than a photosynthetic stimulus. Various
plant photoreceptors, including phytochromes, cryptochromes, and phototropins, initiate cellular
processes in response to specific wavelengths and intensities of radiation. Such responses include
phototropism, shade-avoidance, stomatal opening, and flowering time (Hernández and Kubota,
2017; Li et al., 2012).
19
Blue (B) radiation (400 to 500 nm) has been established as an important driver of
photosynthesis, but can also inhibit extension growth in some species, creating more compact
finished crops (Hernández and Kubota, 2017; Wollaeger and Runkle, 2014, 2015; Zou, 2018).
For instance, Wollaeger and Runkle (2014) reported that the height of impatiens ‘SuperElfin XP
Red’ and salvia ‘Vista Red’ (Salvia splendens) grown under an 18-h photoperiod was 47% to
53% and 41% to 57% shorter, respectively, when grown under sole-source lighting providing
≥25% B radiation when compared to those grown under 100% R radiation. In a separate study,
Zou (2018) reported that geranium ‘Calliope Dark Red’ (Pelargonium interspecific) grown under
100% B radiation was up to 6.7 cm wider in comparison to those grown under 100% R radiation.
Similar effects have been reported for young plants. Wollaeger and Runkle (2015) found that
seedlings of impatiens ‘SuperElfin XP Red’ and salvia ‘Vista Red’ grown under ≥10 µmol∙m–2∙s–
1
of B radiation were 37% to 48% and 29% to 50% shorter than seedlings grown under 100% R
radiation, respectively. Akbarian et al. (2016) found that the stem length of zinnia ‘Art Deco’,
impatiens (Impatiens balsamina), and verbena (Verbena aubletia) seedlings grown under 100%
R radiation was ≈2, ≈7, and ≈4 cm taller in comparison to those grown under 100% B radiation
(Akbarian et al., 2016).
Existing literature suggests that the crop-specific promotion of stem elongation in low-
light environments is mediated in part by responses initiated by photoreceptors such as
phototropins and cryptochromes (Fankhauser and Christie, 2015; Park and Runkle, 2018;
Wollaeger and Runkle, 2015). Although the intensity of all wavebands of radiation decreases
proportionally as the DLI decreases, the specific detection of relatively low intensities of B
radiation (400 to 500 nm) by phototropins can lead to production and distribution of auxins,
which promote cell elongation and apical dominance (Preece and Read, 2005). Additionally,
20
when an insufficient quantity of B radiation is present, cryptochromes cannot act to inhibit
gibberellic acid biosynthesis, as they can under adequate B radiation levels. This leads to cell and
stem elongation as gibberellic acid is synthesized (Runkle and Heins, 2001; Wollaeger and
Runkle, 2015). This stem elongation aids plants in their search for sufficient radiation intensities
(Fankhauser and Christie, 2015; Preece and Read, 2005).
Additionally, B radiation is responsible in part for the regulation of stomatal conductance
(Zeiger, 1984). It has also been successfully utilized as a LD signal for some plants, although it is
ineffective at doing so at low intensities (Meng and Runkle, 2015; Runkle and Heins, 2001; Zou,
2018). For instance, SharathKumar et al. (2021) demonstrated the efficacy of a 4-h DE provided
by 40 µmol·m–2·s–1 of 100% B radiation, creating a 15-h photoperiod, at inhibiting flowering of
greenhouse-grown chrysanthemum ‘Radost’ (Chrysanthemum morifolium; SharathKumar et al.,
2021).
Narrow-band green (G; 500 to 600 nm) light-emitting diodes (LEDs) are not traditionally
utilized in horticultural LED fixtures because they are not as efficient at converting energy into
photons compared to B and R LEDs (Runkle, 2017). Additionally, it was assumed that B and R
radiation has higher photosynthetic efficacy than G radiation, as they are readily absorbed by
chlorophyll a and b while G radiation is not (Kusuma et al., 2020; Liu and van Iersel, 2021).
However, it has been shown that G radiation is useful for photosynthesis, despite a portion of
incoming radiation being reflected by leaves, as photons within this waveband transmit through
leaves and penetrate further into the canopy, (Runkle, 2017). Additionally, recent research has
shown that G radiation is effective at regulating flowering of some photoperiodic crops, although
the photoreceptors responsible for mediating flowering may only absorb relatively small
amounts of it (Meng and Runkle, 2019; Runkle, 2017). For instance, a 16-h photoperiod
21
provided by 13 µmol·m–2·s–1 of DE G radiation delayed time to flower of chrysanthemum
‘Cheryl Spicy Orange’, ‘Cheryl Golden Improved’, and ‘Cheryl Jolly Red’ (Chrysanthemum
morifolium) by 34, 45, and 62 d, respectively, compared to short days (Meng and Runkle, 2019).
Furthermore, G radiation can influence crop morphology, although this is highly dependent on
the flux of G radiation, as well as the flux of other radiation wavebands (Runkle, 2017; Wang
and Folta, 2013). For instance, certain fluxes of G radiation can be perceived by plants as
shading, eliciting a morphogenic response similar to the shade-avoidance response. When
applied in conjunction with FR radiation, subsequent stem elongation was found to be more
substantial than when either of these wavebands were applied alone (Wang and Folta, 2013).
Moreover, it has also been reported that G radiation can inhibit B-radiation induced plant
compaction (Hernández and Kubota, 2017).
Moreover, R radiation is another primary driver of photosynthesis and is effective at
regulating flowering of both SDPs and LDPs at low intensities (Craig and Runkle, 2013; Runkle
and Heins, 2001). FR radiation can influence photosynthesis, but is also very important for
regulating photoperiodic flowering responses, particularly in LDPs. As previously discussed, FR
radiation is important to include in low-intensity photoperiodic lighting in order to elicit or
promote flowering in many LDPs or prevent flowering of many SDPs (Craig and Runkle, 2012,
2013; Walters et al., 2019). For instance, Chrysanthemum ‘Adiva Purple’ generally remained
vegetative when grown under a 4-h NI provided by LED lamps with an R:FR of 0.66 or above.
Furthermore, Dahlia ‘Carolina Burgundy’ flowered when grown under a 4-h NI with any amount
of R radiation, but when grown under SDs or a NI provided by FR radiation alone, only 50% and
40% of plants flowered, respectively (Craig and Runkle, 2013).
22
Additionally, environments devoid of FR radiation can delay flowering of many LDPs.
Runkle and Heins (2001) found that only 88% and 65% of pansy ‘Crystal Bowl Yellow’ (Viola
×wittrockiana) plants grown under an FR radiation deficient filter reached visible flower bud and
flowered, respectively, whereas 100% of plants grown under a neutral filter flowered.
Additionally, a lack of FR radiation delayed flowering of pansy ‘Crystal Bowl Yellow’ by 21 d.
Furthermore, a lack of FR radiation delayed time to visible flower bud by 2 d and 14 d for
campanula ‘Blue Clips’ and coreopsis ‘Early Sunrise’ (Coreopsis ×grandiflora), respectively
(Runkle and Heins, 2001).
Furthermore, SL sources with spectra containing moderate (10-15 µmol·m–2·s–1)
intensities of FR radiation can hasten flowering of many LDPs, compared to SL sources devoid
of FR radiation (Kohler and Lopez, 2021). For instance, snapdragon ‘Liberty Classic Yellow’
grown under SL containing 15 µmol∙m–2∙s–1 of FR radiation for a 16 h·d–1 for 28 d during the
seedling stage flowered 6 d faster than plants grown under SL containing only B and R radiation
during the seedling stage (Kohler and Lopez, 2021).
R and FR radiation is also integral to the onset of the shade-avoidance response. As the
R:FR of the radiation environment decreases, plants typically respond with an extended stem
length and/or leaf size, an effect of the shade-avoidance response (Craig and Runkle, 2013;
Faust, 2011; Hernández and Kubota, 2017, Runkle and Heins, 2001). For instance, SL provided
by LEDs emitting 90 µmol∙m–2∙s–1 of B, G, and R radiation (%; B12G20R68), and additional
fixtures emitting 12 µmol∙m–2∙s–1 of FR radiation, promoted stem elongation of snapdragon
‘Montego Yellow’ seedlings by ≈7 cm in comparison to those grown under SL containing only B
and R radiation (B45R55; Poel and Runkle, 2017). Moreover, Hori et al. (2011) reported that
baby’s breath ‘Bristol Fairy’ (Gypsophila paniculata) cut flowers grown under a 24-h
23
photoperiod created with 16-h of DE lighting providing 20 to 30 µmol∙m–2∙s–1 of 100% FR
radiation or 20 to 30 µmol∙m–2∙s–1 of 100% B radiation expressed 70% and 0% flowering,
respectively, after 18 weeks. Plants grown with 100% FR radiation DE had stems ≈43 cm longer
than those grown under 100% B radiation DE (Hori et al., 2011).
Conclusion
Year-round demand for domestically produced specialty cut flowers continues to
increase. Growers in northern latitudes cannot produce specialty cut flowers outdoors or in high
tunnels during the winter due to inclement temperatures and low radiation intensities and/or
durations. As a result, they must utilize controlled-environment greenhouses to provide more
suitable environmental conditions for growth in order to satisfy the demand for specialty cut
flowers year-round. Further research is needed to determine the responses of specialty cut
flowers to different radiation qualities, quantities, and durations. Such research would prove
useful for growers seeking to diversify their crops and tap into different markets, as well as
smaller, seasonal growing operations seeking to capitalize on the year-round demand of specialty
cut flowers.
24
REFERENCES
25
REFERENCES
Ahmad, I., Rafiq, M.B., Dole, J.M., Abdullah, B., Habib, K., 2017. Production and postharvest
evaluation of selected exotic specialty annual cut flower species in Punjab, Pakistan.
HortTechnology 27, 878–883.
Akbarian, B., Mahna, N., Matloobi, M., 2016. Effects of LED light on seed emergence and
seedling quality of four bedding flowers. J. Ornamental Plants 6, 115–123.
Armitage, A.M., Laushman, J.M., 2003. Specialty cut flowers - the production of annuals,
perennials, bulbs, and woody plants for fresh and dried cut flowers. 2nd ed. Timber Press,
Portland, OR.
Armitage, A.M., Son, K.C., 1992. Shade and photoperiod influence Caryopteris incana used as
cut flowers. HortScience 27,1275–1276.
Armitage, A.M., 1991. Shade affects yield and stem length of field-grown cut flower species.
HortScience 26,1174–1176.
Autio, J., 2000. Supplementary lighting regimes strongly affect the quantity of gerbera flower
yield. Acta Hortic. 515, 91–98.
Bergmann, B., 2019. A brief history of specialty cut flower production. North Carolina State
University Extension, Raleigh, NC. 15 July 2020.
(accessed 15 July 2020).
Blacquière, T., Krijger, D., de Graaf-van der Zande, M., 2002. Photoperiodic lighting of
matricaria (Tanacetum parthenium). Acta Hortic. 580, 117–126.
Blanchard, M.G., Runkle, E.S., 2009. Use of a cyclic high-pressure sodium lamp to inhibit
flowering of chrysanthemum and velvet sage. Scientia Hort. 122, 448–454.
Borthwick, H.A., Hendricks, S.B., 1960. Photoperiodism in plants. Science 132, 1223–1228.
Both, A.J., Faust, J.E., 2017. Light transmission: the impact of glazing material and greenhouse
design. In: Lopez, R.G., Runkle, E.S. (Eds.). Light management in controlled
environments. Meister Media Worldwide, Willoughby, OH, pp. 59–66.
Byczynski, L., 2008. The flower farmer. Chelsea Green Publishing, White River Junction, VT.
Carman, H., 2007. California continues to lead nation in nursery and floral production. CORF
News 11, 1–4.
26
Craig, D.S., Runkle, E.S., 2012. Using LEDs to quantify the effect of the red to far-red ratio of
night-interruption lighting on flowering of photoperiodic crops. Acta Hortic. 956, 179–
186.
Craig, D.S., Runkle, E.S., 2013. A moderate to high red to far-red light ratio from light-emitting
diodes controls flowering of short-day plants. J. Amer. Soc. Hortic. Sci. 138, 167–172.
Craig, D.S., Runkle, E.S., 2016. An intermediate phytochrome photoequilibria from night-
interruption lighting optimally promotes flowering of several long-day plants. Environ.
and Expt. Bot. 121, 132–138.
Damann, M.P., Lyons, R.E., 1993. Juvenility, flowering, and the effects of a limited inductive
photoperiod in Coreopsis grandiflora and C. lanceolata. J. Amer. Soc. Hortic. Sci. 118,
513–518.
Dole, J.M., Warner, R.M., 2017. Lighting greenhouse cut flowers. In: Lopez, R.G., Runkle, E.S.
(Eds.). Light management in controlled environments. Meister Media Worldwide,
Willoughby, OH, pp. 152–158.
Dole, J.M., Wilkins, H.F., 2005. Light. In: Dole, J.M., Wilkins, H.F. (authors). Floriculture
principles and species. 2nd ed. Pearson Education, Upper Saddle River, NJ, pp. 67–78.
Erwin, J.E., Warner, R.M., 2002. Determining the photoperiodic response group and effect of
supplemental irradiance on flowering of several bedding plant species. Acta Hortic. 580,
95–100.
Fankhauser, C., Christie, J.M., 2015. Plant phototropic growth. Current Biol. 25, R384–R389.
Faust, J.E., Dole, J.M., 2021. Cut flowers and foliages. CAB International, Oxfordshire, United
Kingdom.
Faust, J.E., 2011. Light. In: J. Nau (Ed.). Ball Redbook Crop Production: Volume two. 18th ed.
Ball Publishing, Batavia, IL, pp. 83–94.
Faust, J.E., Holcombe, V., Rajapakse, N.C., Layne, D.R., 2005. The effect of daily light integral
on bedding plant growth and flowering. HortScience 40, 645–649.
Floral Marketing Association (FMA), Society of American Florists (SAF), 1996. Recommended
grades & standards for fresh cut flowers. 26 July 2020.
(accessed 26 July 2020).
Goto, T., Muraoka, Y., 2008. Effect of plug transplant age and photoperiod treatment at seedling
stage on cut flower quality of Celosia cristata L. Acta Hortic. 782, 201–206.
27
Granitz, H.M., 2014. Improving the North Carolina cut flower industry: a production and
marketing survey, field and postharvest cut flower variety evaluations, and preemergence
herbicide trials on unlabeled cut flower crops (Masters Thesis). North Carolina State
Univ., Raleigh, NC.
Hammer, P.A., 2011. Photoperiod control. In: C. Beytes (Ed.). Ball Redbook Greenhouses and
Equipment: Volume one. 18th ed. Ball Publishing, Batavia, IL, pp. 52.
Hernández, R., Kubota, C., 2017. Light quality and photomorphogenesis. In: Lopez, R.G.,
Runkle, E.S. (Eds.). Light management in controlled environments. Meister Media
Worldwide, Willoughby, OH, pp. 29.
Hori, Y., Nishidate, K., Nishiyama, M., Kanahama, K., Kanayama, Y., 2011. Flowering and
expression of flowering-related genes under long-day conditions with light-emitting
diodes. Planta 234, 321–330.
Kaiser, C., Ernst, M., 2018. Field-grown specialty cut flowers. Univ. of Kentucky ext. Bul.
CCD-CP-70.
Kelly, J.W., 1991. Field production of cut flowers. HortScience 26,1136–1138.
Knewtson, S.J.B., Janke, R., Kirkham, M.B., Williams, K.A., Carey, E.E., 2010. Trends in soil
quality under high tunnels. HortScience 45,1534–1538.
Köksal, N., Yasemin, S., Özkaya, A., 2017. The effects of photoperiodism on growth and
flowering of marigold (Tagetes erecta). Turkish J. Agr.- Food Sci. Technol. 5, 1189–
1193.
Korczynski, P.C., Logan, J., Faust, J.E., 2002. Mapping monthly distribution of daily light
integrals across the contiguous United States. HortTechnology 12, 12–16.
Kosugi, K., 1962. Studies in production and flowering in gladiolus. Mem. Fac. Agric. Kagawa
Univ. 11.
Kusuma, P., Pattison, P.M., Bugbee, B., 2020. From physics to fixtures to food: current and
potential LED efficacy. Hortic. Res. 56.
Lamont, W.J., 2009. Overview of the use of high tunnels worldwide. HortTechnology 19, 25–29.
Li, J., Terzaghi, W., Deng, X.W., 2012. Genomic basis for light control of plant development.
Protein Cell 3, 106–116.
Liu, J., van Iersel, M.W., 2021. Photosynthetic physiology of blue, green, and red light: light
intensity effects and underlying mechanisms. Front. Plant Sci. 12, 619987.
28
Llewellyn, D., Schiestel, K., Zheng, Y., 2020. Increasing levels of supplemental LED light
enhances the rate flower development of greenhouse-grown cut gerbera but does not
affect flower size and quality. Agronomy 10, 1332.
Lopez, R.G., 2016. Specialty cut flower production and handling.
(accessed 15 July 2020).
Lopez, R.G., Runkle, E.S., 2008. Photosynthetic daily light integral during propagation
influences rooting and growth of cuttings and subsequent development of new guinea
impatiens and petunia. HortScience 43, 2052–2059.
Lopez, R.G., Fisher, P., Runkle, E.S., 2017. Introduction to specialty crop lighting. In: Lopez,
R.G., Runkle, E.S. (Eds.). Light management in controlled environments. Meister Media
Worldwide, Willoughby, OH, pp. 83–94.
Loyola, C.E., Dole, J.M., Dunning, R., 2019a. North American specialty cut flower production
and postharvest survey. HortTechnology 29, 338–359.
Loyola, C.E., Dole, J.M., Dunning, R., 2019b. South and Central America cut flower production
and postharvest survey. HortTechnology 29, 898–905.
Lugasi-Ben-Hamo, M., Kitron, M., Buston, A., Zaccai, M., 2010. Effect of shade regime on
flower development, yield and quality in lisianthus. Scientia Hortic. 124, 248–253.
Mattson, N.S., Erwin, J.E., 2004. The impact of photoperiod and irradiance on flowering of
several herbaceous ornamentals. Scientia Hortic. 104, 275–292.
Meng, Q., Runkle, E.S., 2015. Low-intensity blue light in night-interruption lighting does not
influence flowering of herbaceous ornamentals. Scientia Hortic. 186, 230–238.
Meng, Q., Runkle, E.S., 2019. Regulation of flowering by green light depends on its photon flux
density and involves cryptochromes. Phys. Plantarum 166, 762–771.
Mer, M.S., Attri, B.L., 2015. Effect of photoperiod on flowering in ornamental annuals. J.
Medicinal Plants Studies 3, 121–126.
Moe, R., Grimstad, S.O., Gislerød, H.R., 2006. The use of artificial light in year round
production of greenhouse crops in Norway. Acta Hortic. 711, 35–42.
Montri, A., Biernbaum, J.A., 2009. Management of the soil environment in high tunnels.
HortTechnology 19, 34–36.
Nelson, P.V., 1991. The mechanism of photoperiodism. In: Greenhouse operations and
management. 4th ed. Prentice Hall, Englewood Cliffs, NJ, p. 376.
29
Oh, W., Runkle, E.S., Warner, R.M., 2010. Timing and duration of supplemental lighting during
the seedling stage influence quality and flowering in petunia and pansy. HortScience 45,
1332–1337.
Ohkawa, K., 1987. Growth and flowering of Anemone coronaria ‘de Caen’. Acta Hortic. 205,
159–167.
Olberg, M., Lopez, R.G., 2016. High tunnel and outdoor production of containerized annual
bedding plants in the midwestern United States. HortTechnology 26, 651–656.
Ortiz, M.A., Hyrczyk, K., Lopez, R.G., 2012. Comparison of high tunnel and field production of
specialty cut flowers in the Midwest. HortScience 47, 1265–1269.
Overheim, R.D., Wagner, D.L., 1982. The physical nature of light. In: Light and Color. John
Wiley & Sons, Hoboken, New Jersey, pp. 1–23.
Owen, W.G., Meng, Q., Lopez, R.G., 2018. Promotion of flowering from far-red radiation
depends on the photosynthetic daily light integral. HortScience 53, 465–471.
Owen, W.G., Hilligoss, A., Lopez, R.G., 2016. Late-season high tunnel planting of specialty cut
flowers in the midwestern United States influences yield and stem quality.
HortTechnology 26, 338–343.
Park, Y., Runkle, E.S., 2018. Far-red radiation and photosynthetic photon flux density
independently regulate seedling growth but interactively regulate flowering. Environ.
Expt. Bot. 155, 206–216.
Pazuki, A., Aflaki, F., Pessarakli, M., Gurel, E., Gurel, S., 2017. Plant responses to extended
photosynthetically active radiation (EPAR). Adv. Plants Agric. Res. 2017, 313–318.
Piringer, A.A., Borthwick, H.A., 1961. Effects of photoperiod and kind of supplemental light on
growth, flowering, and stem fasciation of celosia. Amer. J. Bot. 48, 588–592.
Poel, B. R., Runkle, E.S., 2017. Spectral effects of supplemental greenhouse radiation on growth
and flowering of annual bedding plants and vegetable transplants. HortScience 52, 1221–
1228.
Porat, R., Shlomo, E., Halevy, A.H., 1995. Horticultural techniques to improve Celosia plumosa
growth for cut flowers. Scientia Hortic. 63, 209–214.
Pramuk, L.A., Runkle, E.S., 2005. Photosynthetic daily light integral during seedling stage
influences subsequent growth and flowering of Celosia, Impatiens, Salvia, Tagetes, and
Viola. HortScience 40, 1336–1339.
Preece, J.E., Read, P.E., 2005. Plant hormones, In: Preece, J.E., Reed, P.E. (authors). The
biology of horticulture. 2nd ed. John Wiley & Sons, Inc. Hoboken, NJ, pp. 295.
30
Prinzing, D., 2021. New findings on consumer attitudes toward domestic & locally-grown cut
flowers. Slow Flowers Journal. (accessed 27 Apr. 2021).
Produce Marketing Association and Food Marketing Institute (PMAFMI)., 2016. Trends in
mass-market floral. Produce Marketing Association, Newark, DE.
Redman, P.B., Dole, J.M., Maness, N.O., Anderson, J.A., 2002. Postharvest handling of nine
specialty cut flower species. Scientia Hortic. 92, 293–303.
Rüdiger, W., Thümmler, F., 1994. The phytochrome chromophore. In: Hendrick, R.E.,
Kronenberg, G.H.M., (Eds.). Photomorphogenesis in plants. Springer, Berlin, Germany,
pp. 51.
Runkle, E.S., 2018. Causes of flower bud abortion. Greenhouse Prod. News 28, 42.
Runkle, E.S. 2017., Growing plants with green light. Greenhouse Prod. News 27, 58.
Runkle, E.S. 2016., A closer look at far-red radiation. Greenhouse Prod. News 26, 50.
Runkle, E.S., Both, A.J., 2017. Delivering long-day lighting - technology options and costs. In:
Lopez, R.G., Runkle, E.S. (Eds.). Light management in controlled environments. Meister
Media Worldwide, Willoughby, OH, pp. 91–99.
Runkle, E.S., 2013. Using the PGRs Collate and Florel. Greenhouse Prod. News 23, 58.
Runkle, E.S., 2008a. LIP It! Greenhouse Prod. News 18, 66.
Runkle, E.S., 2008b. What is your photoperiod? Greenhouse Prod. News 18, 58.
Runkle, E.S. Heins, R.D., 2006. Manipulating the light environment to control flowering and
morphogenesis of herbaceous plants. Acta Hortic. 711, 51–60.
Runkle, E.S. Heins, R.D., 2001. Specific functions of red, far red, and blue light in flowering and
stem extension of long-day plants. J. Amer. Soc. Hortic. Sci. 126, 275–282.
Runkle, E.S., Lopez, R.G., Fisher, P., 2017. Photoperiodic control of flowering. In: Lopez, R.G.,
Runkle, E.S. (Eds.). Light management in controlled environments. Meister Media
Worldwide, Willoughby, OH, pp. 40.
Sager, J.C., Smith, W.O., Edwards, J.L., Cyr, K.L., 1988. Photosynthetic efficiency and
phytochrome photoequilibria determination using spectral data. Amer. Soc. Agr. Eng. 31,
1882–1889.
31
SharathKumar, M., Heuvelink, E., Marcelis, L.F.M., van Ieperen, W., 2021. Floral induction in
the short-day plant chrysanthemum under blue and red extended long-days. Front. Plant
Sci. 11, 610041.
Sherrard, T., 2011. Supplemental lighting. In: Beytes, C. (Ed.). Ball Redbook Greenhouses and
Equipment: Volume one. 18th ed. Ball Publishing, Batavia, IL, pp. 129.
Shillo, R., Halevy, A.H., 1976. The effect of various environmental factors on flowering of
gladiolus. I. Light intensity. Scientia Hortic. 4, 131–137.
Short, K., Etheredge, C.L., Waliczek, T.M., 2017. Studying the market potential for specialty
cultivars of sunflower cut flowers. HortTechnology 27, 611–617.
Sparks, B.D., 2020. Wanna give cut flowers a go? Here’s what you need to know. Greenhouse
Grower 38, 36.
Styer, R.C., 2002. Using Florel effectively. Greenhouse Prod. News 8, 10–15.
Thomas, B., 2017. Photoperiodism. In: Thomas, B., Murray, B.G., Murphy, D.J. (Eds.).
Encyclopedia of applied plant sciences. 2nd ed. Academic Press Cambrdige, MA, pp.
448–454.
Thomas, B., Vince-Prue, D., 1997. Photoperiodism in plants. 2nd ed. Academic Press, San
Diego, CA.
Torres, A.P., Lopez, R.G., 2012. Measuring daily light integral in a greenhouse. Purdue Univ.
Ext. Bul. HO-238-W.
Torres, A.P., Lopez, R.G., 2011. Photoperiod and temperature influence flowering responses and
morphology of Tecoma stans. HortScience 46, 416–419.
Treder, J., 2003. Effects of supplementary lighting on flowering, plant quality and nutrient
requirements of lily ‘Laura Lee’ during winter forcing. Scientia Hortic. 98, 37–47.
Treder, J., Kubik, M., 2000. Growth, flowering and partitioning of 14C-assimilate in oriental lily
‘Star Gazer’ as affected by light level and flower removal. J. Hortic. Sci. Biotechnol. 75,
277–282.
U.S. Department of Agriculture (USDA), 2019. Floriculture crops 2018 summary.
(accessed 15 July 2020).
U.S. Department of Agriculture (USDA) Natural Resources Conservation Service (NRCS),
2014. Controlling the high tunnel environment.
(accessed 23 July 2020).
32
U.S. International Trade Commission (USITC), 2003. Industry and trade summary cut flowers.
U.S. Intl. Trade Commission, Washington, D.C.
Walters, K.J., Hurt, A.A., Lopez, R.G., 2019. Flowering, stem extension growth, and cutting
yield of foliage annuals in response to photoperiod. HortScience 54, 661–666.
Wang, Y., Folta, K.M., 2013. Contributions of green light to plant growth and development.
Amer. J. Bot. 100, 70–78.
Warner, R.M., 2009. Determination of photoperiod-sensitive stages of development of the short-
day plant celosia. HortScience 44, 328–333.
Whitman, C.M., Heins, R.D., Cameron, A.C., Carlson, W.H., 1998. Lamp type and irradiance
level for daylength extensions influence flowering of Campanula carpatica ‘Blue Clips’,
Coreopsis grandiflora ‘Early Sunrise’, and Coreopsis verticillata ‘Moonbeam’. J. Amer. Soc.
Hortic. Sci. 123, 802–807.
Wien, C.H., 2009. Floral crop production in high tunnels. HortTechnology 19, 56–60.
Wollaeger, H.M., Runkle, E.S., 2015. Growth and acclimation of impatiens, salvia, petunia, and
tomato seedlings to blue and red light. HortScience 50, 522–529.
Wollaeger, H.M., Runkle, E.S., 2014. Growth of impatiens, petunia, salvia, and tomato seedlings
under blue, green, and red light-emitting diodes. HortScience 49, 734–740.
Yue, C., Behe, B.K., 2010. Consumer color preferences for single-stem cut flowers on calendar
holidays and noncalendar occasions. HortScience 45, 78–82.
Yue, C., Hall, C., 2010. Traditional or specialty cut flowers? Estimating U.S. consumers’ choice
of cut flowers at noncalendar occasions. HortScience 45, 382–386.
Zeiger, E., 1984. Blue light and stomatal function. In: Senger, H., (Ed.). Blue light effects in
biological systems. Springer-Verlag Berlin Heidelberg, Heidelberg, Germany, pp. 484.
Zhen, S., van Iersel, M.W., Bugbee, B., 2021. Why far-red photons should be included in the
definition of photosynthetic photons and the measurement of horticultural fixture
efficacy. Front. Plant Sci. 22.
Zhen, S., Bugbee, B., 2020a. Far-red photons have equivalent efficiency to traditional
photosynthetic photons: implications for redefining photosynthetically active radiation.
Plant Cell Environ. 43, 1259–1272.
Zhen, S., Bugbee, B., 2020b. Substituting far-red for traditionally defined photosynthetic photons
results in equal canopy quantum yield for CO2 fixation and increased photon capture
during long-term studies: implications for re-defining PAR. Front. Plant Sci. 11, 581156.
33
Zhen, S., van Iersel, M.W., 2016. Far-red light is needed for efficient photochemistry and
photosynthesis. J. Plant Phys. 209, 115–122.
Zou, Y., 2018. Light-emitting diode lighting quality effects on morphology, growth, flowering,
and carotenoid content of potted plant, cut flower, and medicinal flower production
(Masters Thesis). Univ. of California, Davis, Davis, CA.
34
SECTION 2
35
Daily light integral and/or photoperiod during the young plant and finishing stages influence
floral initiation and quality of witchgrass and marigold cut flowers
Caleb E. Spall and Roberto G. Lopez*
Department of Horticulture, 1066 Bogue Street, Michigan State University, East Lansing, MI
48824-1325, USA
*Corresponding author. Tel.: +1 517-353-0342. E-mail address: rglopez@msu.edu (R.G. Lopez).
36
Abstract
To produce consistent and high-quality specialty cut flowers throughout the year, growers in
temperate climates must utilize controlled environment greenhouses. Research-based
information on photoperiod management and supplemental lighting for specialty cut flowers is
limiting. Therefore, our objectives were 1) to determine the effect of photoperiod during the
young-plant and finishing stages on floral initiation and morphology of witchgrass ‘Frosted
Explosion’ (Panicum capillare) and marigold ‘Xochi’ (Tagetes erecta) and 2) to quantify the
effect of daily light integral (DLI) on floral initiation and morphology of witchgrass during the
finishing stage. Seeds of marigold and multi-pellet seeds of witchgrass were sown and placed
under 9-, 11- (marigold only), 12-, 13-, 14-, 15-, 16-, 18-, or 24-h photoperiods or a 9-h short day
with a 4-h night interruption (NI) from 2200 to 0200 HR. Plugs were distributed among 10-, 11-,
12-, 13-, 14-, 15-, or 16-h photoperiods, or a 4-h NI, for finishing. Witchgrass was finished under
a very low or moderate DLI of ≈3 or 10 mol∙m–2∙d–1, respectively, while marigold was finished
under a DLI of ≈10 mol∙m–2∙d–1. Marigold grown under a photoperiod ≥11 h or a 4-h NI during
the young-plant stage and finished under an 11- or 12-h photoperiod had thick stems and
consistently met the marketable stem length of ≥65 cm. Up to 29 and 107% more stems were
harvestable under 11- and 12-h finishing photoperiods, respectively, compared to a 10-h
finishing photoperiod. Marigold visible buds were delayed, and stems were not harvestable under
photoperiods ≥13 h, or a 4-h NI after 8 weeks. Young witchgrass plants grown under a
photoperiod between 14- and 24-h or a 4-h NI and finished under photoperiods ≥14 h or a 4-h
NI, and at least a moderate DLI, were reliably harvestable (≥50 cm long with a fully developed
panicle). Witchgrass finished under day lengths <13 h (rep. 1) or <14 h (rep. 2) flowered
prematurely and were roughly one-sixth the length of harvestable stems at open flower. All
37
witchgrass stems grown under a very low DLI were shorter and thinner than those grown under a
moderate DLI, and none were harvestable. Therefore, we recommend growing marigold ‘Xochi’
young plants under a photoperiod ≥11h or a 4-h NI for two weeks, and finishing under a 12-h
photoperiod. Additionally, witchgrass ‘Frosted Explosion’ young plants should be grown under a
photoperiod ≥14 h or a 4-h NI for two weeks, and finished under photoperiods ≥14 h, or a 4-h NI
to prevent premature flowering. Witchgrass and Marigold cut flowers should be finished under a
DLI of >10 mol·m–2·d–1 for consistent production of high-quality stems.
Keywords: DLI, light-emitting diodes, Panicum, photoperiodic lighting, specialty cut flowers,
supplemental lighting, Tagetes
Abbreviations: ADT, average daily temperature; B, blue; DE, day extension; DLI, daily light
integral; FR, far red; G, green; LED, light-emitting diode; LD, long day; NI, night interruption;
OF, open flower; PFD, photon flux density; PPFD, photosynthetic photon flux density; R, red;
SD, short day; SDP, short-day plant; SL, supplemental lighting; TOF, time to open flower;
TPFD, total photon flux density; TVB, time to visible flower bud; W, white; VB, visible flower
bud.
38
Introduction
Year-round demand for locally sourced specialty cut flowers continues to increase in the
United States (Faust and Dole, 2021; PMAFMI, 2016). From 2015 to 2018, the number of
domestic cut flower producers with annual sales ≥$100,000 increased by 20%, and producers
reported a wholesale value of $374 million in 2018 (USDA, 2019). Of this, California accounted
for $288 million (77%) of domestic production at least partly because of the coastal climates of
its central and southern counties (Carman, 2007). However, demand persists across the nation
and growers in northern latitudes cannot produce specialty cut flowers outdoors year-round due
to low temperatures and solar radiation during the winter and early spring. Thus, controlled-
environment greenhouses must be utilized to produce high-quality specialty cut flowers year-
round.
Many varieties of specialty cut flowers are categorized as short-day plants (SDPs),
including marigold (Tagetes erecta), celosia (Celosia spp.), and zinnia (Zinnia elegans) (Dole,
2015; Craig and Runkle, 2013). Horticulturally, SDPs are classified as either obligate
(qualitative) or facultative (quantitative), where daylengths equal to or less than a genotype-
specific critical photoperiod are either required for flowering, or accelerate flowering,
respectively (Craig and Runkle, 2013). Young plants with a short day (SD) flowering response
may flower prematurely if grown during periods with natural SDs, resulting in short,
unmarketable stems (Dole and Warner, 2017). Therefore, photoperiodic lighting techniques such
as low-intensity day extension (DE), night interruption (NI), or high-intensity cyclic lighting are
utilized to create long days during the beginning of the production cycle (Meng and Runkle,
2016), ensuring that plants do not flower prematurely, and thereby preventing inferior cut flower
quality (Currey et al., 2013).
39
Preventing premature flowering through photoperiod manipulation may also reduce the
need for plant growth regulator applications. Once flower initiation has occurred, it is rarely
possible to revert plants to a vegetative state by placing them under non-inductive photoperiods
(Runkle, 2008) or by removing flower buds. Thus, flower-aborting plant growth regulators such
as ethephon must be applied, and multiple applications may be necessary over the duration of the
crop cycle (Styer, 2002). Additionally, such plant growth regulators can inhibit internode
elongation and suppress apical dominance (Runkle, 2013), which can be undesirable in some
cases. Therefore, it is recommended that cut flowers be grown under non-inductive photoperiods
for several weeks before flower initiation (Dole and Warner, 2017; Porat et al., 1995).
Limited research-based information detailing photoperiodic lighting applications for
greenhouse-grown SDP cut flowers exists. Blacquière (2002) reported that a low-intensity 2-h NI
was effective at inhibiting flowering of chrysanthemum ‘White Reagen’ and ‘Majoor Bosshardt’
(Chrysanthemum ×morifolium Kitamura) by 28 and 30 d, respectively. Furthermore, Park and
Jeong (2019) demonstrated the efficacy of a 16-h photoperiod and 4-h NI of various light
qualities at inhibiting the flowering of chrysanthemum ‘Gaya Yellow’ for the duration of the
study (46 d) when applied at intensities of 180 and 10 µmol∙m–2∙s–1, respectively, whereas SD
conditions promoted flower bud initiation after 21 d. In addition, LDs and NIs provided by red
(R; 600-700 nm), white (W; 400-700 nm), and far-red (FR; 700-800 nm) radiation resulted in
crops that were 6 to 8 cm taller than those grown under SDs (Park and Jeong, 2019). In a
separate study, pinched celosia ‘Rocket’ (Celosia argentea var. plumosa) grown under a 16-h
photoperiod for 3 weeks, and then an 8-h photoperiod for 29 d, had 4 times as many stems per
plant and were ≈183% taller than pinched plants grown under continuous SDs for 50 d (Porat et
al., 1995).
40
In addition to regulating photoperiod, growers must maintain sufficient radiation
intensities through the use of supplemental lighting (SL) when solar radiation intensities are low
(Wollaeger and Runkle, 2014) to consistently produce high-quality cut flowers. This is especially
important in northern latitudes as the outdoor daily light integral (DLI) can fall to 5 to 10 mol∙m–
2
∙d–1 during the winter and early spring (Korczynski et al., 2002), and can drop further to <5
mol∙m–2∙d–1 in greenhouses due to reflection of incoming radiation from greenhouse glazing and
shading from the greenhouse superstructure (Lopez and Runkle, 2008). Increasing the DLI with
SL to produce greenhouse crops other than specialty cut flowers is well documented. Generally,
a moderate to high DLI (e.g., ≥10 mol∙m–2∙d–1) during the young-plant and finishing stages elicits
favorable growth responses, including a reduction in time to flower and an increase in biomass
and finished plant quality (Faust et al., 2005; Owen et al., 2018; Pramuk and Runkle, 2005).
Research documenting the use of SL to increase the DLI during specialty cut flower production
is limited. By reducing time to flower, growers gain the potential for more production cycles per
season, and thus, the potential for increased annual income. For instance, godetia (Clarkia
amoena) ‘Satin White’, ‘Salmon’, ‘Rose Pink’, and ‘Red’ flowered ≈41, 94, 98, and 114 d faster,
respectively, when grown under SL providing 79 µmol∙m–2∙s–1 from 1800 to 2400 HR in
comparison to those grown without SL in autumn (Anderson, 1993). Although finished stem
length of godetia was 19 to 33% shorter when grown under SL compared to those grown without
SL (Anderson, 1993), the finished stems were still of sufficient length for sale. Similarly, time to
flower and height of oriental lily (Lilium spp.) ‘Laura Lee’ were reduced by an average of 21 d
and 20%, respectively, when grown under SL providing 60 µmol∙m–2∙s–1 for 5 h, compared to
those grown without SL (Treder, 2003).
41
High DLIs also have the potential to increase harvestable cut flower yields (Dole and
Warner, 2017). Stem yield of gerbera ‘Estelle’ and ‘Ximena’ (Gerbera ×cantabrigensis)
increased by 13 and 10 stems, respectively, when grown under a DLI of 6.5 mol∙m–2∙d–1 in
comparison to 3.2 mol∙m–2∙d–1 (Autio, 2000). Stem yield of gerbera ‘Panama’ increased by 40%
as the DLI increased from 5.3 to 11.3 mol∙m–2∙d–1 (Llewellyn et al., 2020). Increased cut flower
yield under higher DLIs is partly due to increased branching for some varieties. Lim et al. (2022)
reported that mountain spike speedwell (Veronica rotunda var. subintegra) and long-leaf spike
speedwell (Veronica longifolia) had 331% and 308% more branches when grown under a DLI of
18.3 mol∙m–2∙d–1 compared to 6.6 mol∙m–2∙d–1 for 12 weeks.
Additional research quantifying the influence of photoperiod and DLI on the growth and
development of greenhouse-grown specialty cut flowers is needed for growers in northern
latitudes. Therefore, the objectives of this study were to 1) determine how various photoperiods
during the young-plant and finishing stages interact to influence floral initiation and morphology
of witchgrass ‘Frosted Explosion’ (Panicum capillare) and marigold ‘Xochi’ (Tagetes erecta),
and 2) quantify how DLI influences floral initiation and morphology of witchgrass during the
finishing stage. We hypothesized that both witchgrass and marigold would exhibit a facultative
SD response, characterized by delayed flowering and longer stem lengths as the young-plant
(seedling stage) and finishing (remainder of the crop cycle after transplant) photoperiods
increased. We also hypothesized that witchgrass grown under a moderate DLI would be of
higher quality, although shorter, compared to those grown under a very low DLI.
42
Materials and methods
Young plant material, culture, lighting treatments, and greenhouse environment
Multi-seed pellets of witchgrass ‘Frosted Explosion’ (PanAmerican Seed, West Chicago,
IL) and seeds of marigold ‘Xochi’ (PanAmerican Seed), were sown in 288-cell (7 mL individual
volume) trays by a commercial propagator (Raker-Roberta's Young Plants, Litchfield, MI).
These varieties were selected as they were recent introductions with reports of premature
flowering. Nine plug trays of witchgrass and ten plug trays of marigold were received on 14 Sep.
2020 (Rep. 1) and 8 Sep. 2021 (Rep. 2), one day after sowing. Each tray was divided into two
blocks of 144 cells. The blocks were randomly and equally distributed in a greenhouse at
Michigan State University (East Lansing, MI; lat. 43 °N) under various photoperiodic treatments.
Photoperiodic treatments consisted of a 9-h SD (0800 to 1700 HR) or a 9-h SD extended with
four R+W+FR light-emitting diode (LED) lamps (Arize™ Greenhouse Pro; General Electric,
Boston, MA) on each bench to create 9-, 11- (marigold only), 12-, 13-, 14-, 15-, 16-, 18-, or 24-h
photoperiods or a 4-h NI from 2200 to 0200 HR. Each LED lamp was covered with multiple
layers of aluminum wire mesh (General purpose aluminum; New York Wire, Grand Island, NY)
to achieve an average total photon flux density (TPFD) of 2 to 3 µmol∙m–2∙s–1 between 400 and
800 nm. The 100-nm waveband ratios (%) emitted by the LED lamps, defined by their B (400-
500 nm), green (G; 500-600 nm), R, and FR photon flux densities (PFDs), were 6:19:45:30.
Young plants were grown in a glass-glazed greenhouse with exhaust fans, evaporative-
pad cooling, radiant hot-water heating, and SL controlled by an environmental control system
(Priva Integro 725; Priva North America, Vineland Station, ON, Canada). High-intensity LED
fixtures (Philips GP-TOPlight DRW-MB; Koninklijke Philips N.V., Eindhoven, Netherlands)
provided a supplemental photosynthetic photon flux density (PPFD) of 120±10 µmol·m–2·s–1 [as
43
measured with a quantum sensor (LI-190R; LI-COR Biosciences, Lincoln, NE)] when the
ambient PPFD dropped below ≈400 µmol·m–2·s–1 between 0800 to 1700 HR. On each bench, a
line quantum sensor (LI-191R, LI-COR, Lincoln, NE) or a quantum sensor (LI-190R, LI-COR,
Lincoln, NE) positioned horizontally at plant height measured PPFD every 10 s and a datalogger
(CR1000; Campbell Scientific, Logan, UT) recorded hourly averages. The actual DLIs during
the young-plant stages of the two replications of the experiment were 10.3 to 11.7 mol·m–2·d–1
(Table 2.1). The 100-nm waveband ratios (%) emitted by the LED fixtures, defined by their B,
G, and R photon flux densities, were 10:5:85.
The greenhouse air average daily temperature (ADT) set point was 20 °C (12 h day/12 h
night at 22/18 °C), with daytime and nighttime temperatures maintained from 0500 to 1700 HR
and 1700 to 0500 HR, respectively. An aspirated thermocouple [36-gauge (0.127 mm diameter)
type E, Omega Engineering, Stamford, CT] positioned in the middle of each bench measured the
air temperature at plant height every 10 s, and the data logger recorded hourly means. The data
logger controlled a 1500-W electric heater underneath each bench to provide supplemental heat
when the nighttime temperature was <19.8 °C. The actual air ADT and average day and night
temperature at plant height of each treatment during the young-plant stages are provided in Table
2.1.
Young plants were irrigated as needed with MSU Plug Special [13N-2.2P-10.8K water-
soluble fertilizer containing (mg·L‒1) 61 nitrogen, 10 phosphorus, 50 potassium, 28.1 calcium,
4.7 magnesium, 1.3 iron, 0.6 manganese, 0.6 zinc, 0.6 copper, 0.4 boron, and 0.1 molybdenum;
(GreenCare Fertilizers Inc., Kankakee, IL)] blended with reverse-osmosis water and applied with
a mist nozzle (Super Fine Fogg-It Nozzle; Fogg-It Nozzle Co. Inc., Belmont, CA).
44
Finished plant lighting treatments, greenhouse environment, and culture
The same high-intensity LED fixtures described above provided a supplemental PPFD of
120±10 µmol·m–2·s–1 (as measured with a quantum sensor) from 0800 to 1700 HR. Additionally,
a combination of whitewash applied to the exterior of the greenhouse (KoolRay Classic Liquid
Shade, Continental Products, Euclid, OH) and shade cloth surrounding benches (Harmony 5120
OE, Ludvig Svensson Inc, Charlotte, NC) was utilized to create DLIs of ≈3 (very low) and ≈10
mol·m–2·d–1 (moderate). The actual DLIs on each bench during the finishing stages of the two
replications of the experiment were calculated and are provided in Tables 2.2 and 2.3. For both
witchgrass and marigold, photoperiods of 10-, 11-, 12-, 13-, 14-, 15-, or 16-h, or a 9-h SD with a
4-h NI from 2200 to 0200 HR, were maintained with the same methods and equipment described
in the last section. Greenhouse temperature set points during the finishing stage were identical to
those in the young-plant stage. The actual air ADT and average day and night temperature at
plant height of each treatment during the finishing stages are provided in Tables 2.2 and 2.3.
One hundred-sixty bulb crates (39.3 cm wide × 59.7 cm long × 17.8 cm tall; 0.23 m2)
were filled with a soilless medium containing (by volume) 70% peat moss, 21% perlite, and 9%
vermiculite (Suremix; Michigan Grower Products Inc., Galesburg, MI). After 14 and 19 d under
young-plant photoperiods for the first rep. (29 Sep. 2020) and second rep. (22 Sep. 2021),
respectively, 160 witchgrass young plants were randomly selected for transplant from the 9-, 12-,
13-, 14-, 16-, 18-, or 24-h young-plant photoperiods or the 4-h NI treatment (1,280 young plants
total). For marigold, 80 young plants were randomly selected for transplant from the 11-, 13-,
14-, 15-, 16-, or 24-h photoperiods or the 4-h NI treatment (560 young plants total). Eight bulb
crates designated for witchgrass seedlings were placed under each photoperiod under the very
45
low and moderate DLI treatments, and four bulb crates designated for marigold seedlings were
placed under each photoperiod under the moderate DLI treatment. Each of the eight or four bulb
crates were divided into two or four sections, respectively, yielding 32 sections total per bench
(16 sections each for witchgrass and marigold). Five witchgrass or marigold seedlings from one
of the aforementioned young-plant treatments were transplanted into a block at a density of 43 or
97 plants per m2, respectively. This was repeated randomly across the sections until 10 seedlings
from each aforementioned witchgrass or marigold young-plant treatment were transplanted per
bench.
One layer of 15 cm supportive netting (HGN32804; Hydrofarm, Petaluma, CA) was
positioned ≈15 cm above the bulb crates on each bench. Plants were irrigated as needed with
MSU Orchid RO Special [13N-1.3P-12.3K water-soluble fertilizer containing (mg∙L‒1) 125
nitrogen, 13 phosphorus, 121 potassium, 76 calcium, 19 magnesium, 1.7 iron, 0.4 copper and
zinc, 0.9 manganese, 0.2 boron, and 0.2 molybdenum; (GreenCare Fertilizers Inc.)] blended with
reverse-osmosis water.
Data collection and analysis
Ten randomly selected young plants from each treatment were monitored daily for the
presence of first visible flower bud (VB). After 14 or 19 d for reps. 1 and 2, respectively, fully-
expanded leaf number, node number, and height from the bottom of the media to the tallest point
of the plant were recorded for these young plants. Additionally, root dry mass (RDM) and shoot
dry mass (SDM) were assessed after gently rinsing media from the roots and drying the plant
material in an oven for a minimum of 3 d at 70 °C.
46
During the finishing stage, the individual stems of each witchgrass plug were monitored
daily for the presence of VB. On this date, node number below the first VB was recorded.
Individual stems of each witchgrass seedling were also monitored daily for the presence of the
first open flower (OF) and the date was recorded. On the date of OF, stem length from the media
to the tallest point of the most developed stem, branch number, and stem caliper at the thickest
point of the stem was recorded with a digital caliper (3-inch carbon fiber digital caliper, General
Tools & Instruments, LLC, New York, NY). For witchgrass, the date of harvest (indicated by
plants becoming ≥50 cm tall with a fully developed panicle) was recorded for the most
developed plant in each plug and for marigold the date of harvest (indicated by plants becoming
≥65 cm tall and terminal blossom 50% open) was recorded for each plant. On the date of harvest,
stem length from the media to the tallest point of the inflorescence, stem caliper at the thickest
point of the stem, branch number, and the total number of initiated inflorescences were recorded
for marigold. Data were analyzed using SAS (version 9.4; SAS Institute, Cary, NC) mixed model
procedure (PROC MIXED) for analysis of variance (ANOVA), and means were separated by
Tukey’s honest significant difference (HSD) test at P ≤0.05. Data from individual reps. were
analyzed separately when interactions between reps. were present.
Results
Young plant morphology and dry mass
Neither witchgrass nor marigold young plant node or leaf number were influenced by
photoperiod (data not reported). Additionally, no plants initiated VBs during the young-plant
stage. However, as the young-plant photoperiod increased from 9 to 16 h, the height of
witchgrass and marigold increased by 0.8 cm and 1.9 cm, respectively, and then decreased over
47
16 h (Fig. 2.1A; 2.1D). As photoperiod increased from 9 to 18 h for witchgrass and 10 to 16 h for
marigold, RDM increased by up to 14 and 52%, respectively. However, as the photoperiod
increased to 24 h, RDM decreased by 9 and 22% for witchgrass and marigold, respectively (Fig.
2.1B; 2.1E). The SDM of witchgrass decreased by 41% as the photoperiod increased from 9 to
24 h. In contrast, as photoperiod increased from 9 to 16 h, the SDM of marigold increased by up
to 32%, after which the SDM decreased by 13% as the photoperiod increased to 24 h (Fig. 2.1C;
2.1F).
Time to visible flower bud
The young-plant and finishing photoperiods interacted to control time to VB (TVB) of
witchgrass (P <.0001). TVB increased quadratically by 18 or 14 d, for rep. 1 and 2, respectively,
when the young-plant photoperiod increased from 9 to 24 h and plants were finished under a 10-
h photoperiod and a moderate DLI (Fig. 2.2A; 2.2C). This relationship further accentuated under
a longer finishing photoperiod; TVB increased quadratically by an average of 38 d as the young-
plant photoperiod increased from 9 to 24 h under a 16-h finishing photoperiod. TVB was also
influenced by finishing photoperiod, particularly as the young-plant photoperiod increased. For
example, TVB of plants grown under a 9-h young-plant photoperiod was delayed by ≈1 d as the
finishing photoperiod increased from 10 to 16 h. However, TVB of plants grown under a 24-h
young-plant photoperiod was delayed by ≈23 d as the finishing photoperiod increased from 10 to
16 h.
Young-plant and finishing photoperiod interacted to influence TVB of marigold during
rep. 1 (P <.0001). However, young-plant photoperiod did not commercially influence TVB and
48
finishing photoperiod had the dominant effect. For instance, TVB of plants finished under 10-h
photoperiods increased by only ≈1 d as the young-plant photoperiod increased from 11 to 24 h
(Fig. 2.3A). TVB increased by only ≈2 d as the young-plant photoperiod increased from 11 to 24
h when plants were finished under a 16-h photoperiod. In comparison, TVB increased by ≈18 d
as the finishing photoperiod increased from 10 to 16 h for plants grown under 11-h young-plant
photoperiods. Conversely, young-plant and finishing photoperiod independently influenced TVB
of plants grown during rep. 2 (P=0.23). As the young-plant photoperiod increased from 11 to 24
h, TVB decreased by ≈1 d (Fig. 2.3B). As the finishing photoperiod increased from 10 to 16 h,
TVB increased by ≈16 d (Fig. 2.3C).
Node number below visible flower bud
Witchgrass seedlings grown under 9- to 12-h or 9- to 13-h photoperiods during rep. 1 and
2, respectively, developed ≈4 nodes below the first VB regardless of finishing photoperiod or
DLI. For plants grown under longer young-plant photoperiods, node number increased
proportionally with the finishing photoperiod. Plants grown under 13- (rep. 1) or 14-h (rep. 2)
young-plant photoperiods had up to ≈2 more nodes below the first VB as the finishing
photoperiod increased from 10 to 16 h, or a 4-h NI, under a moderate DLI. A similar trend was
observed for very-low-DLI grown plants (data not reported). Marigold grown under a 10-h
finishing photoperiod formed VBs after a minimum of six nodes had developed, and node count
increased up to nine nodes as the finishing photoperiod increased to 16 h (data not reported).
49
Time to open flower of witchgrass
Time to open flower (TOF) of witchgrass was influenced by the interaction between the
young-plant and finishing photoperiods, following a trend similar to TVB. TOF increased by up
to 22 or 15 d, for rep. 1 and 2 respectively, as the young-plant photoperiod increased from 9 to
24 h under a finishing photoperiod of 10 h and a moderate DLI (Fig. 2.4A; 2.4C). This effect was
stronger under a longer finishing photoperiod. For instance, under a finishing photoperiod of 16
h, TOF increased quadratically by 38 or 34 d, for rep. 1 and 2 respectively, as the young-plant
photoperiod increased from 9 to 24 h. The effect of finishing photoperiod on TOF accentuated as
the young-plant photoperiod increased. For example, TOF of plants grown under a 12-h young-
plant photoperiod was delayed by ≈5 or 1 d as the finishing photoperiod increased from 10 to 16
h for rep. 1 and 2, respectively. However, flowering of plants grown under a 24-h young-plant
photoperiod was delayed by ≈9 or 19 d as the finishing photoperiod increased from 10 to 16 h for
rep. 1 and 2, respectively. Plants finished under a very low DLI experienced a similar trend,
although fewer plants flowered when grown under ≥13- (rep. 1) or ≥14-h (rep. 2) young-plant
photoperiods, or a 4-h NI, and ≥13- (rep. 1) or ≥14-h (rep. 2) finishing photoperiods or a 4-h NI
(Fig. 2.4B; 2.4D).
Witchgrass stem length, caliper, and branch number at open flower
Stem length of witchgrass at OF was proportional to the TOF and was influenced by the
interaction of young-plant and finishing photoperiods. As the young-plant photoperiod increased
from 9 to 24 h, under a finishing photoperiod of 10 h, stem length increased by an average of
19.5 and 11.0 cm for rep. 1 and 2, respectively (Fig. 2.5A; 2.5C). This effect was strengthened as
50
the finishing photoperiod increased; Stem length of plants finished under a 16-h photoperiod and
grown under a young-plant photoperiod of 9 h was 71.1 and 42.0 cm shorter than those grown
under a 24-h young-plant photoperiod for rep. 1 and 2, respectively. Furthermore, the stem
length of witchgrass increased by 1.0 and 2.0 cm for rep. 1 and 2, respectively, when seedlings
were grown under 9 h photoperiods and the finished plant photoperiod increased from 10 to 16 h.
Conversely, when seedlings were grown under a 24-h young-plant photoperiod and finished
under a 16-h photoperiod, stems were 51.6 or 33.0 cm longer than those finished under a 10-h
photoperiod for rep. 1 and 2, respectively. Similar trends were seen for the very-low-DLI-grown
plants that reached OF, although stem lengths were shorter than the plants finished under the
moderate DLI treatment (Fig. 2.5B; 2.5D). None of the plants finished under the very low DLI
were long enough or developed enough by the end of the study (≈62 d) to be considered
harvestable.
Young-plant and finishing photoperiod interacted to influence stem caliper of witchgrass.
Stem caliper of plants grown under a 10-h finishing photoperiod was 0.8 or 1.0 mm thicker for
rep. 1 and 2, respectively, as the young-plant photoperiod increased from 9 to 24 h (Table 2.4).
This effect accentuated as the finishing photoperiod increased; stem caliper of plants finished
under a 16-h photoperiod was 3.5 or 4.0 mm thicker when the young-plant photoperiod was 24 h
compared to 9 h. Thicker stem calipers were recorded for plants finished under 16-h
photoperiods compared to 10-h photoperiods when young plants were grown under 9-h
photoperiods. This effect strengthened as the young-plant photoperiod increased. Stem caliper of
plants grown under a 24-h young-plant photoperiod was 2.8 or 3.2 mm greater for rep. 1 and 2,
respectively, as the finishing photoperiod increased from 10 to 16 h (Table 2.4). Similar trends,
although attenuated, were seen for the very-low-DLI-grown plants that reached OF. However,
51
stem caliper measurements generally ranged only from 0.4 to 2.3 mm for rep. 1 and 1.2 to 2.5
mm for rep. 2. Plants grown under a moderate DLI had 1 to 3 branches at OF, regardless of
young-plant or finishing photoperiod, while those grown under a very low DLI had 0 to 2
branches at OF.
Time to harvest
During rep. 1, witchgrass stems were only harvestable when seedlings were grown under
a photoperiod ≥13 h and finished under a photoperiod ≥13 h and a moderate DLI. Plants finished
under photoperiods <13 h flowered prematurely and were unmarketable. Generally, plants grown
under 13-h young-plant and finishing photoperiods became harvestable the fastest, whereas those
grown under a NI during the young-plant and finishing stages were the slowest to reach harvest
(Table 2.5). During rep. 2, plants grown under young-plant photoperiods <14 h and finishing
photoperiods <14 h flowered prematurely. All harvestable plants were harvested within a 10 to
13-d timeframe, depending on rep. (Table 2.5). Plants finished under a very low DLI did not
yield harvestable stems.
Only marigold grown under 10- to 12-h finishing photoperiods were harvestable by the
end of the study (≈50 d). However, up to 29 and 107% more stems were harvested under 11- and
12-h finishing photoperiods, respectively, compared the 10-h photoperiod (data not reported).
Time to harvest of marigold finished under 10-, 11-, and 12-h photoperiods ranged from 40 to 48
d after transplant (Table 2.6).
52
Marigold stem length, caliper, branch and inflorescence number at harvest
Finishing photoperiod had the dominant effect on marigold stem length at harvest, and
young-plant and finishing photoperiods did not interact to influence length of the stems that
became harvestable. As the finishing photoperiod increased from 10 to 12 h, stem length at
harvest increased linearly from 69 to 74 cm. While plants grown under photoperiods >12 h were
not harvestable at the end of the study, they were at least 65 cm long, regardless of finishing
photoperiod (data not reported), indicating the potential for all stems to eventually reach
marketability. Stem caliper and branch and inflorescence number at harvest of plants grown
under finishing photoperiods of 10 to 12 h were not significantly different (data not reported).
Discussion
The results of this study further support that photoperiod manipulation during the young-
plant and finishing stages, in addition to maintaining or increasing the DLI, can aid in producing
high-quality specialty cut flowers while reducing crop time. Growers can manipulate these
environmental parameters to improve finished cut flower quality and reduce time to harvest.
These techniques are particularly useful when the natural photoperiod is not conducive with the
photoperiodic responses of the crop to be grown, or when solar radiation is limiting.
When grown under inductive conditions, specialty cut flowers can flower prematurely
with unmarketable stems (Dole and Warner, 2017). Witchgrass demonstrated this phenomenon,
which is consistent with several other publications on photoperiodic lighting of SDPs. Plants in
this study grown under 9- to 12-h (rep. 1) or 9- to 13-h (rep. 2) young-plant photoperiods
flowered prematurely, regardless of finishing photoperiod. Premature flowering was also seen
53
for witchgrass finished under 10- to 12-h (rep. 1) or 9- to 13-h (rep. 2) finishing photoperiods,
regardless of young-plant photoperiod. These findings indicate that from a horticultural
standpoint, the critical photoperiod for flowering of witchgrass ‘Frosted Explosion’ is 12 to 13 h.
During rep. 2, stem lengths of plants grown under photoperiods ≤13 h were only 10 to 32 cm
long, which is below the market minimum of 50 cm (BloomStudios, 2020, personal
communication). Similarly, marigold finished under inductive photoperiods were shorter than
those under non-inductive photoperiods, although all treatments would have yielded marketable
stem lengths upon flowering.
Jensen et al. (2012) reported similar trends to witchgrass after investigating the
photoperiodic response of Amur silvergrass (Miscanthus sacchariflorus), a grass used as a
biofuel. Flowering of Miscanthus was delayed by 83 d under LDs (15.3-h photoperiods)
compared to gradually decreasing SDs (15.3-h photoperiods for 21 d, followed by 119 d of a
decreasing photoperiod consistent with that at 34.1 °N), designating it as a facultative SDP.
Plants grown under LDs accumulated ≈52% more biomass (stem and leaf tissue) compared to
plants grown under SDs, aligning with the witchgrass stem caliper increase in the present study.
The authors hypothesized that stem length under LDs would likely have been longer than those
grown under SDs if their experiment ran longer, as Miscanthus exhibits rapid stem elongation
during the emergence of flag leaves, which is ≈18 d after floral initiation (Jensen et al., 2012).
However, the experiment was terminated before LD-mediated plant elongation would have
occurred. During rep. 2 of the present study, witchgrass finished under a 14-h LD were up to
224% longer at OF than plants finished under a 13-h SD, suggesting a similar stem elongation
response.
54
In the present study, witchgrass may have a similar sensitivity to inductive photoperiods
as celosia (Celosia argentea var. plumosa) during the young-plant stage. Warner (2009) reported
that the SDP celosia ‘Gloria Scarlet’ overcame juvenility and perceived inductive treatments 9 to
12 d after cotyledon emergence. Plants exposed to twelve 9-h SDs before cotyledon emergence,
then placed under a 4-h NI from 2200 to 0200 HR, had ≈7 fewer nodes below the terminal
inflorescence than plants grown under continuous LDs. Moreover, plants exposed to 12 SDs
beginning 3 d after cotyledon emergence had a similar node number below the terminal
inflorescence compared to those under continual SDs (Warner, 2009). This could explain why
plants grown under inductive conditions during the young-plant stage flowered prematurely,
even when transferred to non-inductive finishing conditions. Further experimentation to
determine when witchgrass begins reproductive development may be necessary. However,
growers should avoid premature flower-inductive conditions to ensure proper market
specifications can be met.
TVB of marigold was negligibly influenced by young-plant photoperiod, suggesting
marigold was not induced to flower during the first two weeks of growth. This is inconsistent
with Warner (2006), who identified the photoperiod-sensitive stages of the SDPs cosmos ‘Sonata
White’ (Cosmos bipinnatus) and signet marigold ‘Tangerine Gem’ (Tagetes tenuifolia). It was
reported that both species were receptive to inductive conditions after 1 to 2 leaf pairs had
unfolded, with five 9-h SDs delivered after cotyledon emergence promoting flowering of cosmos
by 23 d compared to a constant 4-h NI treatment. Furthermore, marigold exposed to 5 SDs after
cotyledon emergence flowered ≈10 d faster than plants grown under continual LDs (Warner,
2006).
55
Although the variables interacted, finishing photoperiod had a greater effect on TVB of
marigold than young-plant photoperiod. Plants finished under 10- to 12-h photoperiods had
faster TVB than those finished under photoperiods ≥13 h (Fig. 2.3C). Therefore, horticulturally
speaking, the critical photoperiod for flowering of marigold ‘Xochi’ was determined to be 12 h.
Marigold’s flowering responses align with other studies detailing photoperiodic responses of
SDPs (Kang et al., 2019; Park et al., 2013). TVB of zinnia ‘Dream Land’ decreased by 10 d
when finished under 9-h SDs instead of 4-h NI (Park et al., 2013). Similarly, Kang et al. (2019)
demonstrated 8-h SDs promoted flowering of kalanchoe ‘Lipstick’ (Kalanchoe blossfeldiana),
while 16-h photoperiods or 4-h NIs inhibited flowering, regardless of NI TPFDs. Unlike
marigold ‘Xochi’, kalanchoe ‘Lipstick’ was ≈45 or 33% taller under SDs compared to plants
grown under a 16-h LD or a 4-h NI, respectively (Kang et al., 2019). However, this may have
been due to the absence of stem elongation associated with flowering, as plants grown under LDs
or NIs did not transition from vegetative to reproductive growth.
Absorbed radiation is the driving force for photosynthesis and subsequent plant growth
and development. Thus, SL must be utilized when solar radiation is limiting to produce high-
quality cut flowers year-round. SL had a substantial effect on the growth and development of
witchgrass. TVB and TOF was hastened for plants grown under a moderate DLI compared to
those grown under a very low DLI. Similarly, Faust et al. (2005) reported that time to flower of
vinca ‘Pacific Lilac’ (Catharanthus roseus) and zinnia ‘Dreamland Rose’ decreased by three and
10 d, respectively, when grown under a DLI of 43 mol∙m–2∙d–1 compared to 5 mol∙m–2∙d–1. In a
separate study, jasmine tobacco ‘Domino White’ (Nicotiana alata Link & Otto) and helipterum
(Helipterum roseum Hook.) flowered 17 and 6 d faster, respectively, when grown under SL
56
providing 50 µmol∙m–2∙s–1 for 18 h compared to those grown without SL (Erwin and Warner,
2002).
100% of witchgrass plants grown under young-plant photoperiods between 13 and 24 h
or a 4-h NI (rep. 1) or 14 and 24 h or a 4-h NI (rep. 2), and finished under photoperiods ≥13 h, or
a 4-h NI (rep. 1) or ≥14 h, or a 4-h NI (rep. 2), and a moderate DLI yielded harvestable stems.
Conversely, no plants finished under a low DLI yielded harvestable stems. Likewise, Furufuji et
al. (2014) reported that cut rose ‘Tint’ (Rosa spp.) yield was 101% higher when grown with a
supplemental DLI of 5.8 mol∙m–2∙d–1 compared to those grown without SL. Moreover,
greenhouse-grown lisianthus ‘Echo Champagne’ and ‘Rosita White’ (Eustoma spp.) produced 12
and 2 more stems per m2, respectively, when grown under 67% shade compared to 88% shade
for 5 weeks (Lugasi-Ben-Hamo et al., 2010). Furthermore, witchgrass stem length, caliper, and
branch number improved when grown under moderate DLIs. Similarly, Torres and Lopez (2011)
found that stem caliper of yellow trumpet bush ‘Mayan Gold’ (Tecoma stans) seedlings
increased by 133% as the DLI increased from 0.8 to 25.2 mol∙m–2∙d–1. In another study, stem
caliper and height of mountain spike speedwell (Veronica rotunda var. subintegra) increased by
110% and 77%, respectively, as the DLI increased from 3.6 to 18.3 mol∙m–2∙d–1 (Lim et al.,
2022).
In conclusion, the present study indicates that high-quality marigold ‘Xochi’ cut flowers
can be produced in a timely fashion when young plants are grown under any photoperiod
between 11 and 24 h, or a 4-h NI, and finished under a 12-h photoperiod, as stem yield was
highest under this finishing photoperiod compared to 10- or 11-h finishing photoperiods. The
interactions between young-plant and finishing photoperiods were not commercially impactful
for the stems of marigold that reached VB or were harvestable. While marigold finished under
57
photoperiods >12 h did not reach harvestability during the study, they all developed flower buds
and were likely to have become harvestable, after a delay, compared to marigold finished under
photoperiods <13 h. As such, the influence of young-plant and finishing photoperiod was not
empirically quantified for these plants and could be investigated further in another study.
Moreover, high-quality witchgrass ‘Frosted Explosion’ cut flowers can be grown under any
photoperiod between 14 and 24 h, or a 4-h NI, during the young-plant stage, and finished under
photoperiods equal to or greater than 14-h, or a 4-h NI, to prevent premature flowering and
subsequent inferior quality. While these photoperiods yielded cut flowers of similar thickness,
16-h photoperiods can be maintained to produce longer witchgrass stems. Witchgrass should be
grown under at least a moderate DLI of ≥10 mol·m–2·d–1 during the finishing stage to produce
cut flowers with sufficient stem lengths and calipers for market. Growers once limited to
producing witchgrass and marigold outdoors in temperate seasons may use these
recommendations to produce these varieties in greenhouses during the winter and early spring,
allowing for consistent production.
Acknowledgements
We gratefully acknowledge BloomStudios and The Association of Specialty Cut Flower
Growers for providing funding and supplies. We also gratefully acknowledge Nate DuRussel,
John Gove, and Alec Fowler for greenhouse assistance and data collection, Hydrofarm for
netting, Ludvig Svensson for shade cloth, Raker Roberta’s Young Plants for sowing seeds,
Syndicate Sales for floral supplies, and Signify for LED supplemental lighting fixtures. This
work was supported by the USDA National Institute of Food and Agriculture, Hatch project
MICL02472.
58
APPENDIX
59
Table 2.1. Actual average daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], air average
daily temperature (ADT), day temperature, and night temperature [mean ± SD (°C)] throughout
the duration of the witchgrass and marigold young-plant stage for reps. 1 and 2.
Photoperiod DLI ADT Day Night
(h) (mol·m–2·d–1) (°C) (°C) (°C)
Rep. 1
9 –z 20.1 ± 1.0 21.6 ± 2.5 18.6 ± 2.5
11 –z 20.1 ± 1.0 21.6 ± 2.5 18.6 ± 2.5
12 –z –z –z –z
13 –z –z –z –z
14 10.6 ± 3.6 21.5 ± 1.0 22.8 ± 2.1 20.1 ± 2.0
15 11.1 ± 1.9 –z –z –z
16 10.9 ± 3.7 21.4 ± 1.1 23.1 ± 3.0 19.7 ± 3.1
18 10.5 ± 3.6 21.2 ± 0.9 22.8 ± 2.7 19.5 ± 2.8
24 10.7 ± 3.8 22.0 ± 1.3 23.7 ± 3.9 20.3 ± 4.3
4-h NI 10.3 ± 3.2 21.0 ± 1.0 22.0 ± 1.8 19.9 ± 1.5
Rep. 2
9 10.5 ± 5.4 21.1 ± 2.0 21.5 ± 4.1 20.6 ± 3.5
11 –z 20.4 ± 2.1 21.0 ± 4.1 19.7 ± 3.1
12 11.3 ± 6.2 20.6 ± 2.1 21.3 ± 4.3 19.8 ± 3.1
13 10.6 ± 4.9 20.5 ± 2.3 21.0 ± 4.1 19.8 ± 3.2
14 –z 20.4 ± 2.2 20.9 ± 4.0 19.8 ± 3.1
15 10.4 ± 6.7 20.9 ± 2.5 21.4 ± 4.4 20.4 ± 3.7
16 11.7 ± 6.4 20.4 ± 2.3 20.9 ± 4.1 19.8 ± 3.2
18 10.8 ± 5.6 19.8 ± 2.1 19.8 ± 3.5 19.9 ± 3.3
24 10.7 ± 5.9 20.5 ± 2.0 20.8 ± 3.7 20.2 ± 3.2
4-h NI 11.6 ± 5.9 21.1 ± 2.1 22.0 ± 5.0 20.1 ± 3.3
z
No data recorded
60
Table 2.2. Actual average daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], air average daily temperature (ADT), mean day
temperature, and mean night temperature [mean ± SD (°C)] throughout the duration of the witchgrass finishing stage for reps. 1 and 2.
Moderate DLI Very low DLI
Photoperiod DLI ADT Day Night DLI ADT Day Night
(h) (mol·m–2·d–1) (°C) (°C) (°C) (mol·m–2·d–1) (°C) (°C) (°C)
Rep. 1
10 10.0 ± 4.8 19.7 ± 1.4 22.1 ± 3.3 17.2 ± 3.2 2.9 ± 1.0 20.3 ± 1.2 22.3 ± 2.4 18.3 ± 2.4
11 –z 19.9 ± 1.2 22.0 ± 2.7 17.9 ± 2.5 2.7 ± 1.0 20.6 ± 0.9 22.7 ± 2.3 18.5 ± 2.1
12 10.1 ± 4.8 20.9 ± 1.2 23.3 ± 3.0 18.5 ± 2.9 2.7 ± 0.9 20.3 ± 1.1 22.1 ± 2.2 18.5 ± 2.2
13 –z –z –z –z 2.9 ± 0.9 20.4 ± 1.0 22.2 ± 2.3 18.5 ± 2.1
z
14 10.6 ± 4.9 20.8 ± 1.3 23.0 ± 3.2 18.6 ± 2.5 – 19.9 ± 1.3 21.9 ± 2.6 17.9 ± 2.7
15 11.6 ± 4.7 –z –z –z 3.1 ± 0.9 20.7 ± 0.9 22.6 ± 2.6 18.8 ± 2.2
16 10.8 ± 4.4 20.3 ± 1.0 22.2 ± 2.8 18.4 ± 2.4 2.9 ± 1.1 20.6 ± 1.0 22.6 ± 2.2 18.6 ± 2.0
4-h NI 10.3 ± 3.9 20.4 ± 0.9 21.8 ± 2.2 19.0 ± 1.7 –z 20.3 ± 1.4 22.4 ± 2.6 18.1 ± 2.6
Rep. 2
10 10.0 ± 3.5 20.1 ± 1.8 22.2 ± 3.1 18.0 ± 3.2 3.0 ± 1.6 19.7 ± 1.6 21.8 ± 2.6 17.6 ± 3.2
11 9.8 ± 3.1 20.1 ± 1.3 22.2 ± 2.7 18.0 ± 2.8 3.6 ± 1.8 19.6 ± 1.2 22.0 ± 2.5 17.3 ± 2.7
12 10.0 ± 3.2 20.3 ± 1.1 22.3 ± 2.7 18.4 ± 2.5 3.4 ± 1.8 19.8 ± 1.9 21.9 ± 3.0 17.7 ± 3.2
13 –z 19.1 ± 2.0 20.6 ± 3.8 17.1 ± 3.7 2.9 ± 1.8 19.6 ± 1.2 21.9 ± 2.5 17.3 ± 2.7
14 10.1 ± 3.0 19.1 ± 1.9 20.6 ± 3.8 17.3 ± 3.4 3.1 ± 2.0 19.4 ± 1.8 21.6 ± 2.9 17.2 ± 3.5
15 9.8 ± 2.9 20.4 ± 1.6 22.4 ± 2.7 18.4 ± 3.2 3.2 ± 1.7 19.5 ± 1.4 21.6 ± 2.5 17.3 ± 2.9
16 10.2 ± 4.2 –z –z –z 2.9 ± 1.4 19.5 ± 1.6 21.6 ± 2.5 17.4 ± 3.2
4-h NI –z 18.9 ± 1.9 20.9 ± 2.9 16.8 ± 3.4 3.6 ± 1.9 19.2 ± 1.4 21.5 ± 2.5 17.0 ± 3.0
Z
No data recorded
61
Table 2.3. Actual daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], air average daily temperatures (ADTs), mean day
temperature, and mean night temperature [mean ± SD (°C)] throughout the duration of the marigold finishing stage for reps. 1 and 2.
Photoperiod DLI ADT Day Night
(h) (mol·m–2·d–1) (°C) (°C) (°C)
Rep. 1
10 10.6 ± 4.2 19.5 ± 1.4 22.1 ± 3.5 17.0 ± 3.3
11 –z 20.1 ± 1.3 22.1 ± 2.7 18.0 ± 2.6
12 10.6 ± 4.8 21.0 ± 1.2 23.4 ± 3.0 18.7 ± 2.9
13 –z –z –z –z
14 11.0 ± 4.8 21.0 ± 1.3 23.2 ± 3.2 18.8 ± 2.5
15 11.9 ± 4.6 –z –z –z
16 11.2 ± 4.4 20.4 ± 1.1 22.3 ± 2.9 18.4 ± 2.5
4-h NI 10.8 ± 3.8 20.5 ± 0.9 21.9 ± 2.2 19.1 ± 1.7
Rep. 2
10 9.8 ± 2.0 19.8 ± 1.9 22.0 ± 3.3 17.7 ± 3.5
11 9.0 ± 1.7 20.4 ± 1.3 22.3 ± 2.6 18.6 ± 2.7
12 10.0 ± 2.1 20.4 ± 1.2 22.3 ± 2.8 18.6 ± 2.7
13 9.8 ± 1.5 19.3 ± 2.1 20.6 ± 4.1 17.6 ± 4.0
14 10.1 ± 2.6 19.4 ± 2.0 20.5 ± 4.1 17.9 ± 3.6
15 10.0 ± 1.7 20.8 ± 1.5 22.7 ± 2.7 18.9 ± 3.2
16 10.4 ± 2.9 –z –z –z
4-h NI –z 19.1 ± 2.2 20.8 ± 3.3 17.2 ± 3.8
Z
No data recorded
62
Table 2.4. Effects of young-plant photoperiod (9, 12, 13, 14, 16, 18, 24 h, or a 4-h NI) and finishing photoperiod (10, 11, 12, 13, 14,
15, 16 h, or a 4-h NI) on stem caliper (mm) of witchgrass ‘Frosted Explosion’ (Panicum capillare) at open flower. Cut flowers were
finished under a moderate DLI of ≈10 mol∙m–2∙d–1.
Young-plant photoperiod (h)
Finishing 9 12 13 14 16 18 24 NI
photoperiod
(h)
Rep. 1
10 0.61 1.05 1.31 1.17 1.68 1.41 1.43 1.41
11 0.69 0.97 2.00 2.06 2.35 2.35 2.41 2.08
12 0.74 1.61 2.36 2.26 2.58 2.50 2.61 2.70
13 0.58 2.66 3.77 4.10 3.59 4.01 4.11 3.64
14 0.82 3.06 3.96 4.62 4.30 4.19 3.73 4.05
15 0.63 3.22 3.98 4.61 3.91 4.35 4.43 3.94
16 0.67 3.59 4.45 4.40 4.12 4.38 4.25 3.89
NI 0.88 3.37 3.89 4.38 3.96 3.93 4.23 4.04
Rep. 2
10 1.42 1.74 2.34 2.31 2.33 2.63 2.39 2.39
11 1.53 1.82 2.40 2.54 2.61 2.66 2.52 2.53
12 1.65 1.74 2.43 2.26 2.80 2.78 2.68 2.46
13 1.53 1.89 2.58 3.16 3.25 3.47 3.46 3.46
14 1.60 1.88 3.37 5.30 5.61 5.10 5.29 4.88
15 1.64 1.97 3.20 4.83 5.56 5.18 5.41 5.17
16 1.56 1.95 3.63 5.75 5.75 5.91 5.61 5.40
NI 1.73 1.87 3.86 5.28 5.91 5.41 5.62 5.28
63
Table 2.5. Effects of young-plant photoperiod (9, 12, 13, 14, 16, 18, 24 h, or a 4-h NI) and finishing photoperiod (10, 11, 12, 13, 14,
15, 16 h, or a 4-h NI) on time to harvest (d) from the date of transplant of witchgrass ‘Frosted Explosion’ (Panicum capillare) grown
under a moderate DLI of ≈10 mol∙m–2∙d–1.
Young-plant photoperiod (h)
Finishing 9 12 13 14 16 18 24 NI
photoperiod (h)
Rep. 1
10 –z –z –z –z –z –z –z –z
11 –z –z –z –z –z –z –z –z
12 –z –z –z –z –z –z –z –z
13 –z –z 53 53 53 51 53 54
14 –z –z 54 53 53 54 56 56
15 –z –z 58 57 56 58 53 55
16 –z –z 55 55 57 58 58 55
NI –z –z 60 55 55 56 56 58
Rep. 2
10 –z –z –z –z –z –z –z –z
11 –z –z –z –z –z –z –z –z
12 –z –z –z –z –z –z –z –z
13 –z –z –z –z –z –z –z –z
14 –z –z –z 49 48 47 47 48
15 –z –z –z 51 49 53 51 51
16 –z –z –z 52 50 48 48 51
NI –z –z –z 60 58 57 57 60
z
No harvestable plants by end of study
64
Table 2.6. Effects of young-plant photoperiod (11, 13, 14, 15, 16, or 24 h, or a 4-h NI) and finishing photoperiod (10, 11, 12, 13, 14,
15, 16 h, or a 4-h NI) on time to harvest (d) from the date of transplant of marigold ‘Xochi’ (Tagetes erecta).
Young-plant photoperiod (h)
Finishing 11 13 14 15 16 24 NI
photoperiod (h)
Rep. 1
10 46 48 43 47 44 44 44
11 41 43 43 44 43 45 40
12 43 46 42 43 46 46 45
13 –z –z –z –z –z –z –z
14 –z –z –z –z –z –z –z
15 –z –z –z –z –z –z –z
16 –z –z –z –z –z –z –z
NI –z –z –z –z –z –z –z
Rep. 2
10 45 44 42 45 42 43 43
11 42 41 41 42 44 44 42
12 44 44 41 41 42 42 41
13 –z –z –z –z –z –z –z
14 –z –z –z –z –z –z –z
15 –z –z –z –z –z –z –z
16 –z –z –z –z –z –z –z
NI –z –z –z –z –z –z –z
z
No harvestable plants by end of study
65
Table 2.7. Regression analysis equations and adjusted R2 for height; root dry mass; and shoot dry mass in response to photoperiod (P;
9-, 11-, 12-, 13-, 14-, 15-, 16-, 18-, 24-h photoperiods or a 4-h NI) of marigold ‘Xochi’ (Tagetes erecta) or witchgrass ‘Frosted
Explosion’ (Panicum capillare). All models are in the form of: ƒ = y0 + a*P + b*P2.
Parameter y0 a b R2
Marigold ‘Xochi’
Height (cm) 6.09z 0.74 -0.02 0.231
Root dry mass (g) -0.00 0.00 -5.14E-05 0.164
Shoot dry mass (g) -0.01 0.00 0.00 0.092
Witchgrass ‘Frosted Explosion’
Height (cm) 5.14 0.38 -0.01 0.037
Root dry mass (g) 0.00 0.00 -1.92E-05 0.056
Shoot dry mass (g) 0.02 -5.27E-06 -6.83E-06 0.050
z
Coefficients for model equations were used to generate Figs. 2.1A through 2.1F.
66
Table 2.8. Regression analysis equations and adjusted R2 for time to visible flower bud; time to open flower; and stem length at open
flower in response to young-plant photoperiod (YP; 9, 12, 13, 14, 16, 18, or 24 h, or a 4-h night interruption; NI) and finishing
photoperiod (FP; 10, 11, 12, 13, 15, or 16 h, or a 4-h NI) of witchgrass ‘Frosted Explosion’ (Panicum capillare). All models are in the
form of: ƒ = y0 + a*YP + b*FP + c*YP2 + d*FP2+e*(YP*FP). Cut flowers were finished under a moderate DLI of ≈10 mol∙m–2∙d–1 or
a very low DLI of ≈3 mol∙m– 2∙d– 1.
Parameter y0 a b c d e R2
Time to visible flower bud (d)
Rep. 1 Moderate DLI -100.65z 7.86 5.09 -0.29 -0.18 0.25 0.772
y
Rep.1 Very low DLI 11.17 -13.33 -0.29 0.64 0.636
Rep. 2 Moderate DLI 4.56 -7.53 -0.23 0.20 0.36 0.759
Rep. 2 Very low DLI 7.67 -10.27 -0.18 0.49 0.589
Time to open flower (d)
Rep. 1 Moderate DLI 9.05 -9.41 -0.30 0.36 0.19 0.650
Rep.1 Very low DLI 10.26 -9.07 -0.27 0.40 0.742
Rep. 2 Moderate DLI 4.72 -5.98 -0.21 0.16 0.29 0.739
Rep. 2 Very low DLI 6.70 -6.39 -0.16 0.29 0.582
Stem length at open flower (cm)
Rep. 1 Moderate DLI 8.17 -13.32 -0.41 0.43 0.62 0.781
Rep.1 Very low DLI -96.37 5.42 9.84 -0.14 -0.32 0.682
Rep. 2 Moderate DLI 10.08 -13.50 -0.24 0.70 0.587
Rep. 2 Very low DLI -90.87 2.91 12.44 -0.07 -0.43 0.472
z
Coefficients for model equations were used to generate Figs. 2.2, 2.4, and 2.5
y
Blank cells = 0
67
Table 2.9. Regression analysis equations and adjusted R2 for time to visible flower bud in response to young-plant photoperiod (11,
13, 14, 15, 16, 24 h, or a 4-h NI) and/or finishing photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) of marigold ‘Xochi’ (Tagetes
erecta). Models 3A is in the form of: ƒ = y0 + a*YP + b*FP + c*YP2 + d*FP2 and models 3B and 3C are in the form of: ƒ = y0 + a*P
+ b*P2.
Figure y0 a b c d R2
3A -94.22z 3.26 11.72 0.09 -0.33 0.631
y y
3B 40.34 -1.36 0.03 0.004
y y
3C -57.28 10.05 -0.26 0.697
z
Coefficients for model equations were used to generate Figs. 2.3A through 2.3C
y
Blank cells = 0
68
Marigold 'Xochi' Witchgrass 'Frosted Explosion'
14 9
A D
13
8
12
Height (cm) Height (cm)
11
7
10
9 6
0 0
0.012 B 10 12 14 16 18 20 22 24 28
E 10 12 14 16 18 20 22 24 28
0.008
0.011
Root dry mass (g) Root dry mass (g)
0.010
0.007
0.009
0.008
0.006
0.007
0.006 0.005
0.000 0.000
0.050
C 10 12 14 16 18 20 22 24 28
F 10 12 14 16 18 20 22 24 28
0.024
Rep 2
Shoot dry mass (g) Shoot dry mass (g)
0.045 0.022
0.020
0.040
0.018
0.035
0.016
0.030
0.000 0.000
10 12 14 16 18 20 22 24 NI
28 10 12 14 16 18 20 22 24 NI
28
Photoperiod (h) Photoperiod (h)
Figure 2.1. Effect of 9, 11, 12, 13, 14, 15, 16, 18, 24 h photoperiods or a 4-h night interruption
(NI) on the height (A; D), root dry mass (B; E), and shoot dry mass (C; F) of marigold ‘Xochi’
(Tagetes erecta) and witchgrass ‘Frosted Explosion’ (Panicum capillare) young plants. Black
symbols indicate means; error bars indicate standard error of the mean. NI means were excluded
from regressions. Figure 1-F presents data from replication 2 as trends from replication 1 were
not significant. Coefficients are presented in Table 2.7.
69
Figure 2.2. Effects of young-plant photoperiod (9, 12, 13, 14, 16, 18, 24 h, or a 4-h NI) and
finishing photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to visible flower bud of
witchgrass ‘Frosted Explosion’ (Panicum capillare). Figures represent (A) moderate-DLI-grown
(≈10 mol∙m–2∙d–1) cut flowers from replication 1, (B) very-low-DLI-grown (≈3 mol∙m–2∙d–1) cut
flowers from replication 1, (C) moderate-DLI-grown cut flowers from replication 2, and (D)
very-low-DLI-grown cut flowers from replication 2. Black symbols represent individual data
points for sequential photoperiods; red symbols represent means from NI treatments. NI means
were excluded from response surfaces. Model predictions are represented by response surfaces;
coefficients are presented in Table 2.8.
70
Figure 2.3. Effect of young-plant photoperiod (11, 13, 14, 15, 16, 24 h, or a 4-h NI) and/or
finishing photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to visible flower bud
(TVB) of marigold ‘Xochi’ (Tagetes erecta). Figures represent (A) the interaction between
young-plant and finishing photoperiod on TVB of plants from replication 1, (B) the effect of
young-plant photoperiod on TVB of plants from replication 2, and (C) the effect of finishing
photoperiod on TVB of plants from replication 2. In figure 3A, black symbols represent
individual data points for sequential photoperiods; red symbols represent means from NI
treatments. NI means were excluded from response surfaces and regressions. Coefficients are
presented in Table 2.9.
71
Figure 2.4. Effects of young-plant photoperiod [9, 12, 13, 14, 16, 18, 24 h, or a 4-h night
interruption (NI)] and finishing photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to
open flower of witchgrass ‘Frosted Explosion’ (Panicum capillare). Figures represent (A)
moderate-DLI-grown (≈10 mol∙m–2∙d–1) cut flowers from replication 1, (B) very-low-DLI-grown
(≈3 mol∙m–2∙d–1) cut flowers from replication 1, (C) moderate-DLI-grown cut flowers from
replication 2, and (D) very-low-DLI-grown cut flowers from replication 2. Black symbols
represent individual data points for sequential photoperiods; red symbols represent means from
NI treatments. NI means were excluded from response surfaces. Model predictions are
represented by response surfaces; coefficients are presented in Table 2.8.
72
Figure 2.5. Effects of young-plant photoperiod [9, 12, 13, 14, 16, 18, 24 h, or a 4-h night
interruption (NI)] and finishing photoperiod (10, 11, 12, 13, 14, 15, 16 h, or a 4-h NI) on stem
length of witchgrass ‘Frosted Explosion’ (Panicum capillare) at open flower. Figures represent
(A) moderate-DLI-grown (≈10 mol∙m–2∙d–1) cut flowers from replication 1, (B) very-low-DLI-
grown (≈3 mol∙m–2∙d–1) cut flowers from replication 1, (C) moderate-DLI-grown cut flowers
from replication 2, and (D) very-low-DLI-grown cut flowers from replication 2. Black symbols
represent individual data points for sequential photoperiods; red symbols represent means from
NI treatments. NI means were excluded from response surfaces. Model predictions are
represented by response surfaces; coefficients are presented in Table 2.8.
73
REFERENCES
74
REFERENCES
Ahmad, I., Rafiq, M.B., Dole, J.M., Abdullah, B., Habib, K., 2017. Production and postharvest
evaluation of selected exotic specialty annual cut flower species in Punjab, Pakistan.
HortTechnology 27, 878–883.
Anderson, G., 1993. Supplemental lighting for cut godetia production for Christmas and
Valentine’s Day crops. Univ. of Kentucky, Lexington, KY (Accessed May 25, 2022).
Autio, J., 2000. Supplementary lighting regimes strongly affect the quantity of gerbera flower
yield. Acta Hortic. 515, 91–98.
Blacquière, T., 2002. How much light is needed for the prevention of flowering of cut
chrysanthemums when using high intensity HPS lighting as a night break. Acta Hortic.
580, 69–75.
Blanchard, M.G., Runkle, E.S., 2009. Use of a cyclic high-pressure sodium lamp to inhibit
flowering of chrysanthemum and velvet sage. Scientia Hortic.122, 448–454.
Carman, H., 2007. California continues to lead nation in nursery and floral production. CORF
News 11, 1–4.
Craig, D.S., Runkle, E.S., 2013. A moderate to high red to far-red ratio from light-emitting
diodes controls flowering of short-day plants. J. Amer. Soc. Hortic. Sci.138, 167–172.
Currey, C.J., Lopez, R.G., Runkle, E.S., 2013. Managing photoperiod in the greenhouse. Purdue
Univ. Ext. Bul. HO-253-W.
Dole, J.M., 2015. Photoperiod responses of ten specialty cut flowers. Acta Hortic. 1097, 163–
168.
Dole, J.M., Warner, R.M., 2017. Lighting greenhouse cut flowers. In: Lopez, R.G., Runkle, E.S.
(Eds.). Light management in controlled environments. Meister Media Worldwide,
Willoughby, OH, pp. 152–158.
Erwin, J.E., Warner, R.M., 2002. Determining the photoperiodic response group and effect of
supplemental irradiance on flowering of several bedding plant species. Acta Hortic. 580,
95–100.
Faust, J.E., Dole, J.M., 2021. Cut flowers and foliages. Oxfordshire, United Kingdom: CAB
International.
Faust, J.E., Holcombe, V., Rajapakse, N.C., Layne, D.R., 2005. The effect of daily light integral
on bedding plant growth and flowering. HortScience 40, 645–649.
75
Furufuji, S., Amaki, W., Fukui, H., 2014. Cut rose production under supplemental lighting with
super bright white light emitting diodes. Acta Hortic. 1055, 383–386.
Jensen, E., Robson, P., Norris, J., Cookson, A., Farrar, K., Donnison, I., Clifton-Brown, J., 2012.
Flowering induction in the bioenergy grass Miscanthus sacchariflorus is a quantitative
short-day response, whilst delayed flowering under long days increases biomass
accumulation. J. of Exp. Bot. 64, 541–552.
Kang, D.I., Jeong, H.K., Park, Y.G., Jeong, B.R., 2019. Flowering and morphogenesis of
kalanchoe in response to quality and intensity of night interruption light. Plants 8.
Korczynski, P.C., Logan, J., Faust, J.E., 2002. Mapping monthly distribution of daily light
integrals across the contiguous United States. HortTechnology 12, 12–16.
Lim, S.H., Im, N.H., An, S.K., Lee, H.B., Kim, K.S., 2022. Daily light integral affects
photosynthesis, growth, and flowering of Korean native Veronica rotunda and V.
longifolia. Hortic. Environ. Biotechnol.
Llewellyn, D., Schiestel, K., Zheng, Y., 2020. Increasing levels of supplemental LED light
enhances the rate flower development of greenhouse-grown cut gerbera but does not
affect flower size and quality. Agronomy 10, 1332.
Lopez, R.G., Runkle, E.S., 2008. Photosynthetic daily light integral during propagation
influences rooting and growth of cuttings and subsequent development of new guinea
impatiens and petunia. HortScience 43, 2052–2059.
Lugasi-Ben-Hamo, M., Kitron, M., Buston, A., Zaccai, M., 2010. Effect of shade regime on
flower development, yield and quality in lisianthus. Scientia Hortic. 124, 248–253.
Meng, Q., Runkle, E.S., 2016. Control of flowering using night-interruption and day-extension
LED lighting. In: Kozai, T., Fujiwara, K., Runkle, E.S. (Eds.). LED lighting for urban
agriculture. Springer Singapore, Singapore, pp. 191–201.
Owen, W.G., Meng, Q., Lopez, R.G., 2018. Promotion of flowering from far-red radiation
depends on the photosynthetic daily light integral. HortScience 53, 465–471.
Park, Y.J., Kim, Y.J., Kim, K.S., 2013. Vegetative growth and flowering of Dianthus, Zinnia,
and Pelargonium as affected by night interruption at different timings. Hortic. Environ.
Biotechnol. 54, 236–242.
Park, Y.G., Jeong, B.R., 2019. Night interruption light quality changes morphogenesis,
flowering, and gene expression in Dendranthema grandiforum. Hortic. Environ. and
Biotechnol. 60, 167–173.
Porat, R., Shlomo, E., Halevy, A.H., 1995. Horticultural techniques to improve Celosia plumosa
growth for cut flowers. Scientia Hortic. 63, 209–214.
76
Pramuk, L.A., Runkle, E.S., 2005. Photosynthetic daily light integral during seedling stage
influences subsequent growth and flowering of Celosia, Impatiens, Salvia, Tagetes, and
Viola. HortScience 40, 1336–1339.
Produce Marketing Association and Food Marketing Institute (PMAFMI; 2016). Trends in mass-
market floral. Produce Marketing Association, Newark, DE.
Runkle, E.S., 2008. LIP It! Greenhouse Prod. News 18, 66.
Runkle, E.S., 2013. Using the PGRs Collate and Florel. Greenhouse Prod. News 23, 58.
Styer, R.C., 2002. Using Florel effectively. Greenhouse Prod. News 8, 10–15.
Torres, A.P., Lopez, R.G., 2011. Photoperiod and temperature influence flowering responses and
morphology of Tecoma stans. HortScience 46, 416–419.
Treder, J., 2003. Effects of supplementary lighting on flowering, plant quality and nutrient
requirements of lily ‘Laura Lee’ during winter forcing. Scientia Hortic. 98, 37–47.
U.S. Department of Agriculture (USDA; 2019). Floriculture crops 2018 summary.
https://downloads.usda.library.cornell.edu/usda-
esmis/files/0p0966899/rr1728124/76537c134/floran19.pdf (accessed 15 July 2020).
Warner, R.M., 2009. Determination of photoperiod-sensitive stages of development of the short-
day plant celosia. HortScience 44, 328–333.
Warner, R., 2006. Using limited inductive photoperiod for scheduling Cosmos bipinnatus and
Tagetes tenuifolia. Acta Hortic. 711, 267–272.
Wollaeger, H.M., Runkle, E.S., 2014. Low daily light integrals in northern latitudes.
https://www.canr.msu.edu/news/low_daily_light_integrals_in_northern_latitudes
(accessed 30 November 2020).
77
SECTION 3
78
Daily light integral, but not photoperiod, commercially influences time to flower and finished
quality of dianthus specialty cut flowers
Caleb E. Spall and Roberto G. Lopez*
Department of Horticulture, 1066 Bogue Street, Michigan State University, East Lansing, MI
48824-1325, USA
*Corresponding author. Tel.: +1 517-353-0342. E-mail address: rglopez@msu.edu (R.G. Lopez).
79
Abstract
Due to burgeoning year-round demand, greenhouse growers across the U.S. are
increasingly interested in producing specialty cut flowers year-round for local and regional
markets. However, outdoor or high tunnel production is not possible year round in northern
latitudes due to low temperatures and radiation intensities experienced during the winter and
early spring. Additionally, natural short days in these seasons can limit what photoperiodic crops
can be grown. Thus, our objectives were to quantify the influence of photoperiod and daily light
integral (DLI) on greenhouse-grown dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose
Magic’ (Dianthus barbatus interspecific) cut flowers during the young-plant and finishing
stages. Seeds of both cultivars were placed under 9-, 10-, 11-, 12-, 13-, 15-, or 16-h photoperiods
and a DLI of either ≈5 or 10 mol·m–2·d–1 one day after sowing. After four weeks, seedlings from
several young-plant photoperiods were distributed across 11-, 12-, 13-, 14-, 15-, or 16-h
photoperiods, or a 4-h night interruption (NI), under a DLI of either ≈5 (low) or 14 (moderate)
mol·m–2·d–1 for finishing. Young-plant photoperiod generally had a statistical, but not
commercial, influence on development and finished cut flower quality, while a 16-h finishing
photoperiod marginally hastened development compared to an 11-h finishing photoperiod.
Additionally, stems were 11 to 13 cm longer when finished under a 16-h photoperiod compared
to those under an 11-h photoperiod. Daylength minimally influenced time to flower and harvest,
indicating a day-neutral flowering response. However, plants finished under a moderate DLI
reached VB and were harvestable 9 to 10 d earlier than those finished under a low DLI.
Additionally, ≈99% of cut flowers finished under a moderate DLI were harvestable, while only
up to 32% and 57% of dianthus ‘Amazon Rose Magic’ and ‘Amazon Neon Cherry’ finished
under a low DLI became harvestable. While finished stem lengths were comparable between
DLI treatments, cut flower stems were up to 29.6% thicker under a moderate DLI. These
findings indicate that high-quality greenhouse-grown dianthus ‘Amazon Neon Cherry’ and
‘Amazon Rose Magic’ cut flowers can be produced when grown under any photoperiod between
9 and 16 h for four weeks (during the young-plant stage), and finished under any photoperiod
80
between 11 and 16 h, or a 4-h NI, during finishing. If longer stems are desired, plants can be
finished under a 16-h photoperiod. Young plants should be grown under a moderate DLI of ≥10
mol∙m–2∙d–1 to promote biomass accumulation. Additionally, plants should be finished under a
moderate DLI of ≥14 mol·m–2·d–1 to reduce crop time and increase stem thickness and yield.
Keywords: light-emitting diodes, photoperiodic lighting, supplemental lighting, DLI, long-day
plants, plugs, greenhouse
Abbreviations: ADT, average daily temperature; B, blue; DE, day extension; DLI, daily light
integral; FR, far-red; G, green; LEDs, light-emitting diodes; HPS, high-pressure sodium; LDPs,
long-day plants; NI, night interruption; OF, open flower; PAR, photosynthetically active
radiation; PFD, photon flux density; PPFD, photosynthetic photon flux density; R, red; SL,
supplemental lighting; TOF, time to open flower; TPFD, total photon flux density; TTH, time to
harvest; TVB, time to visible flower bud; W, white; VB, visible flower bud.
81
Introduction
The demand for locally-grown specialty cut flowers persists across the United States
throughout the year (Faust and Dole, 2021; PMAFMI, 2016), though steady production is
typically limited to regions with consistently high temperatures and radiation intensities.
California dominated specialty cut flower sales in 2018, accounting for 77% of domestic
wholesale value (USDA, 2019), due to its coastal climates and overall suitability for outdoor
production year-round (Carman, 2007). Conversely, northern regions of the United States
experience low temperatures and radiation intensities during the winter and early spring,
preventing production of specialty cut flowers outdoors or in high tunnels year-round. However,
controlled-environment greenhouses allow for the production of these crops during these
seasons, giving the growers potential to capitalize on burgeoning local demand throughout the
year.
Many specialty cut flowers, including ageratum (Ageratum houstonianum), snapdragon
(Antirrhinum majus), and stock (Matthiola incana) are characterized as long-day plants (LDPs;
Currey et al., 2011). Horticulturally, LDP flowering responses are classified as either obligate
(qualitative) or facultative (quantitative), where daylengths at or longer than a genotype-specific
critical photoperiod are required for, or hasten flowering, respectively (Currey et al., 2011).
During production of LDP cut flowers, growers must employ photoperiodic lighting while
growing plants that flower in response to long days (LDs) during seasons when the day length is
naturally short to induce flowering when desired. Low-intensity day-extension (DE) and night-
interruption (NI) lighting, in addition to high-intensity cyclic lighting, can be utilized to create
LDs in the greenhouse (Runkle and Both, 2017).
82
It is recommended that cut flowers be grown under non-inductive photoperiods before
inducing flowering to prevent early flowering and short stem lengths (Dole and Warner, 2017;
Cavins and Dole, 2001). If flower induction occurs prematurely, growers may need to apply
flower-aborting plant growth regulators (PGRs) such as ethephon to abort undesirable flower
buds (Styer, 2002), as it is rarely possible to revert crops to a vegetative state by changing the
day length to non-inductive photoperiods (Runkle, 2008). Moreover, such PGRs can reduce stem
lengths (Runkle, 2013), which can be undesirable for cut flower growers in some cases as longer
stems are generally desired by florists and consumers alike (Ahmad et al., 2017). Thus, short
days (SDs) should be maintained during the early stages of LDP crop cycles to prevent
premature flowering, and to reduce or eliminate the need for subsequent PGR applications.
Controlling the photoperiod to regulate LDP flowering is well documented. In one study,
snapdragon ‘Spring Giants Mix’ (Antirrhinum majus) exhibited a facultative LDP response, as
time to anthesis was slowed by 19 d when grown under an 8-h photoperiod, compared to a 16-h
photoperiod provided by DE lighting (Dole, 2015). Plants grown under a 16-h photoperiod had
13 less nodes below the first flower compared to those grown under an 8-h photoperiod (Dole,
2015). Moreover, Heins and Wilkins (1977) reported that dianthus ‘White Sim’ (Dianthus
caryophyllus) eventually flowered when grown under a constant 8-h SD, but flowering was
accelerated by 88 d when plants were grown under a constant LD created with a 5-h NI
beginning at 2200 HR. Erwin and Warner (2002) designated dianthus ‘Ideal Cherry Picotee’
(Dianthus chinensis) as a facultative LDP because flowering occurred under both 8-h SDs and
LDs created with a 4-h NI, though plants grown under SDs developed 2 to 3 more leaves below
the first OF than those grown under LDs.
83
Moreover, Blacquière et al. (2002) induced flowering of the obligate LDPs matricaria
‘Snowball’ and ‘Single Vegmo’ (Tanacetum parthenium) under 7- to 8-h DE lighting creating an
18-h photoperiod, as well as under NI lighting applied at an intensity of 2 µmol∙m–2∙s–1
beginning at various points throughout the dark period. Matricaria ‘Snowball’ and ‘Single
Vegmo’ flowered after ≈50 d when a 2-h NI was initiated 8 h into the skotoperiod. However,
flowering was delayed by ≈6 d under a 14-h photoperiod created with 2 h of DE lighting
beginning before dawn, compared to the former treatment (Blacquière et al., 2002).
In addition to photoperiod, radiation intensity must be accounted for when cultivating cut
flowers in northern latitudes during the winter and early spring. The photosynthetic daily light
integral (DLI) outdoors can be as low as 5 to 10 mol·m–2·d–1 during these seasons and can be
further reduced to ≤5 mol·m–2·d–1 within greenhouses due to poor weather conditions and
shading from the superstructure (Korczynski et al., 2002; Lopez and Runkle, 2008). Reductions
in radiation intensity can lengthen production cycles, cause flower bud abortion, and reduce cut
flower stem yield and finished quality (Dole and Warner, 2017; Marcelis et al., 2006). Marcelis
et al. (2006) reported that a 1% reduction in radiation intensity can reduce cut rose (Rosa spp.)
and cut chrysanthemum (Chrysanthemum ×morifolium) yield by 0.4% to 1.2% and 0.3% to
1.0%, respectively. Moreover, this relationship acts strongest during the winter when radiation
levels are typically low; a minor decrease in radiation intensity while radiation levels are high
have a lesser effect on yield (Marcelis et al., 2006).
The effects of radiation intensity on specialty cut flower growth, development, and
quality have been documented for some varieties. The average fresh mass and time to flower of
oriental lily ‘Star Gazer’ (Lilium spp.) grown without supplemental lighting (SL) was reduced by
22% and flowering was delayed by 22 d, respectively, compared to those grown under 60
84
µmol∙m–2∙s–1 of SL provided by HPS lamps for 10 h∙d–1 (Treder and Kubik, 2000). Furthermore,
lisianthus ‘Echo Champagne’ (Eustoma grandiflorum) produced 12 more stems per m2 with an
average of 1 more bud per stem when grown under 67% shade for 5 weeks, in comparison to
plants grown under 88% shade for 5 weeks (Lugasi-Ben-Hamo et al., 2010). Likewise, lisianthus
‘Rosita White’ produced 2 more stems per m2 with an average of 3 more buds per stem when
grown under 67% shade for 5 weeks, compared to plants grown under 88% shade for 5 weeks
(Lugasi-Ben-Hamo et al., 2010). In a separate study, the time between flower bud initiation and
harvest of gerbera ‘Ultima’ (Gerbera jamesonii) decreased by 4 d as the total DLI increased
from 5.3 to 11.3 mol∙m–2∙d–1 (Llewellyn et al., 2020). Additionally, flower diameter and stem
yield of gerbera ‘Panama’ increased by 11% and 40%, respectively, as the total DLI increased
from 5.3 to 11.3 mol∙m–2∙d–1 (Llewellyn et al., 2020).
Day length manipulation and sufficient radiation intensities are critical for greenhouse
production of high-quality specialty cut flowers. More research regarding the influence of the
light environment for greenhouse-grown specialty cut flower production would be beneficial to
growers in northern latitudes who are seeking to cultivate such crops year-round. Therefore, the
objectives of this study were to quantify the influence of photoperiod and DLI on greenhouse-
grown dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’ (Dianthus barbatus
interspecific) cut flowers during the young-plant and finishing stages. These cultivars were
chosen due to their facultative LDP classification (PanAmerican Seed, 2021), and popularity as
filler cut flowers. We hypothesized that longer day lengths during finishing would yield shorter
plants than those finished under shorter photoperiods, as LDs would induce flowering
prematurely. Additionally, we expected that both cultivars tested would demonstrate a facultative
LDP response during the finishing stage. Furthermore, we hypothesized that cut flower length
85
would increase under very-low DLIs compared to moderate DLIs, and that moderate DLIs would
yield more harvestable plants with thicker stems.
Materials and methods
Young plant material, culture, lighting treatments, and greenhouse environment
Seeds of dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’, two established
specialty cut flower varieties, (PanAmerican Seed, West Chicago, IL), were sowed in 288-cell (7
mL individual volume) trays by a commercial propagator (Raker-Roberta's Young Plants,
Litchfield, MI). Trays were received on 01 January 2021 [replication (Rep.) 1] and 10 January
2022 (Rep. 2), one day after sowing. Young plants were grown in a glass-glazed greenhouse at
Michigan State University (East Lansing, MI; lat. 43 °N) with exhaust fans, evaporative-pad
cooling, radiant hot-water heating, and supplemental lighting controlled by an environmental
control system (Priva Integro 725; Priva North America, Vineland Station, ON, Canada). Upon
receipt, each tray was divided into two blocks of 144 cells. Blocks were randomly distributed
under various photoperiods and a moderate (≈10 mol·m–2·d–1) or low (≈5 mol·m–2·d–1) DLI
treatment. Photoperiod treatments consisted of a 9-h SD or a 9-h SD extended with four
red+white+far-red (R+W+FR) light-emitting diode (LED) lamps (Arize™ Greenhouse Pro;
General Electric, Boston, MA) on each bench to create a 9-, 10-, 11-, 12-, 13-, 15-, or a 16-h
photoperiod under either DLI treatment. Each LED lamp was covered with multiple layers of
aluminum wire mesh (General purpose aluminum; New York Wire, Grand Island, NY) to
achieve an average total photon flux density (TPFD) of 2 to 3 µmol∙m–2∙s–1 between 400 and 800
nm. The 100-nm waveband ratios (%) emitted by the R+W+FR LED lamps, defined by their
86
blue (B; 400-500 nm), green (G; 500-600 nm), R (600-700 nm), and FR radiation (700-800 nm)
photon flux densities (PFDs), were 4:17:49:29.
LED fixtures (Philips GP-TOPlight DRW-MB; Koninklijke Philips N.V., Eindhoven,
Netherlands) provided a supplemental photosynthetic photon flux density (PPFD) of 175 ± 10
µmol·m–2·s–1 [as measured with a light quantum sensor (LI-190R; LI-COR Biosciences, Lincoln,
NE)] from 0800 to 1700 HR on the south side of the greenhouse. The 100-nm waveband ratios
(%) emitted by the LED fixtures, defined by their B, G, and R PFDs, were 10:5:85. Shade cloth
(Harmony 5120 OE, Ludvig Svensson Inc, Charlotte, NC) was stretched above benches to create
a low DLI of 5 mol·m–2·d–1. On each bench, a line quantum sensor (LI-191R, LI-COR, Lincoln,
NE) or a quantum sensor (LI-190R, LI-COR, Lincoln, NE) positioned horizontally at plant
height measured PPFD every 10 s and a datalogger (CR1000; Campbell Scientific, Logan, UT)
recorded hourly averages. The actual DLIs during the young-plant stages of the two replications
were calculated and are provided in Table 3.1.
The greenhouse air average daily temperature (ADT) set point was 20 °C (day/night
22/18 °C), with day temperatures maintained from 0500 to 1700 HR and night temperatures
maintained from 1700 to 0500 HR. An aspirated thermocouple [36-gauge (0.127-mm diameter)
type E, Omega Engineering, Stamford, CT] positioned in the middle of each bench measured the
air temperature at plant height every 10 s, and the datalogger recorded hourly means. The data
logger controlled a 1500-W electric heater underneath each bench to provide supplemental heat
when the nighttime temperature was <19.8 °C. The actual air ADTs and average day and night
temperatures at plant height of each treatment during the young-plant stages during the two reps
were calculated and are provided in Table 3.1.
87
Young plants were irrigated as needed with MSU Plug Special [13N–2.2P–10.8K water-
soluble fertilizer containing (mg·L‒1) 61 nitrogen, 10 phosphorus, 50 potassium, 28.1 calcium,
4.7 magnesium, 1.3 iron, 0.6 manganese, 0.6 zinc, 0.6 copper, 0.4 boron, and 0.1 molybdenum;
(GreenCare Fertilizers Inc., Kankakee, IL)] blended with reverse-osmosis water and applied with
a mist nozzle (Super Fine Fogg-It Nozzle; Fogg-It Nozzle Co. Inc., Belmont, CA).
Finished plant lighting treatments, greenhouse environment, and culture
The same high-intensity LED fixtures described above were utilized to provide a DLI of
≈5 or 14 mol·m–2·d–1. The actual DLI of each treatment during the finishing stages of the two
reps of the experiment were calculated and are provided in Table 3.2. Photoperiods of 11-, 12-,
13-, 14-, 15-, or 16-h, or a 4-h NI from 2200 to 0200 HR were maintained under each DLI with
the same methods and equipment described above. Greenhouse temperature set points during the
finishing stage were identical to those in the young-plant stage. The actual air ADT and average
day and night temperatures at plant height of each treatment during the finishing stages during
the two reps. were calculated and are provided in Table 3.2.
One hundred-sixty-eight bulb crates (39.3 cm wide × 59.7 cm long × 17.8 cm tall; 0.23
m2) were filled with a soilless medium containing (by volume) 70% peat moss, 21% perlite, and
9% vermiculite (Suremix; Michigan Grower Products Inc., Galesburg, MI). Twelve bulb crates
were placed under each photoperiod/DLI treatment. Each bulb crate was divided into two blocks,
yielding 24 total blocks.
After 28 d under photoperiod treatments, 112 young plants each of moderate-DLI-grown
dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’ were selected from the 9-, 11-, 12-,
13-, 15-, or 16-h photoperiods (672 young plants total of each cultivar). Four plugs of each
88
cultivar from one of the aforementioned young-plant treatments were transplanted into an
individual block at a density of 32 plants per m2. This was repeated randomly across 12 blocks
until 8 seedlings from each aforementioned young-plant treatment were transplanted per bench.
One layer of 15 cm supportive netting (HGN32804; Hydrofarm, Petaluma, CA) was
positioned ≈15 cm above the bulb crates on each bench. Plants were irrigated as needed with
MSU Orchid RO Special [13N–1.3P–12.5K water-soluble fertilizer containing (mg∙L‒1) 125
nitrogen, 13 phosphorus, 121 potassium, 76 calcium, 19 magnesium, 1.7 iron, 0.4 copper and
zinc, 0.9 manganese, 0.2 boron, and 0.2 molybdenum; (GreenCare Fertilizers Inc.)] blended with
reverse-osmosis water.
Data collection and analysis
During the young-plant stage, 10 young plants per treatment per cultivar were monitored
daily for the presence of visible flower bud (VB). Dianthus ‘Amazon’ cut flowers exhibit an
“off-type” phenotype at a rate of 3% to 5%. Off-type young plants are characterized by
stretching ≈14 d after sowing and premature flowering, which decreases crop uniformity and
overall quality (BloomStudios, 2020, personal communication). The number of off-type young
plants was recorded, in addition to their date of identification. Off-type young plants were
discarded upon identification. After 28 d under treatments, fully-expanded leaf number, height
from the top of the media to the tallest point of the plant, and node number were recorded for
vegetative plugs. Additionally, root dry mass (RDM) and shoot dry mass (SDM) were assessed
after gently rinsing media from the roots and drying the plant material in an oven for a minimum
of 3 d at 70 °C.
89
During the finishing stage, plants were monitored daily for the presence of VBs. The date of
first VB was recorded. On the date of VB, node number and leaf number below the first VB were
recorded. Plants were also monitored daily for the presence of the first open flower (OF). The
date of first OF for each plant was recorded. Additionally, the date of harvest (at stage 3 and ≥50
cm with a full flower head; BloomStudios, 2020, personal communication) was recorded for
each plant. On the date of harvest, harvestable stem length to the tallest point of the
inflorescence, total number of initiated inflorescences, and branch number were recorded, and
stem caliper was measured at the thickest point of the stem with a digital caliper (3-inch carbon
fiber digital caliper, General Tools & Instruments, LLC, New York, NY). Data were analyzed
using SAS (version 9.4; SAS Institute) mixed model procedure (PROC MIXED) for ANOVA
and means were separated by Tukey’s HSD test at P ≤ 0.05. Data from individual reps were
analyzed separately when interactions between reps were present.
Results
Young plant morphology, dry mass, and off-type incidence
Young plant height increased as the photoperiod increased from 9 to 16 h for both
cultivars tested, regardless of DLI treatment (Fig. 3.1A; 3.1D; 3.2A; 3.2D). Dianthus ‘Amazon
Neon Cherry’ and ‘Amazon Rose Magic’ young plants were 1.5 to 2.0 cm taller when grown
under a 16-h photoperiod compared to a 9-h photoperiod under moderate or low DLIs,
respectively. Moreover, young plants of both cultivars grown under a low DLI were of
comparable height to those grown under a moderate DLI.
SDM was not commercially influenced by photoperiod for either cultivar tested (Fig.
3.1B; 3.1E; 3.2B; 3.2E). SDM of dianthus 'Amazon Neon Cherry' and 'Amazon Rose Magic' was
90
up to 45% and 54% greater, respectively, when plants were grown under a moderate DLI
compared to a low DLI. RDM was 19% lower for dianthus ‘Amazon Rose Magic’ young plants
grown under a 16-h photoperiod compared to those under a 9-h photoperiod and a moderate DLI
(Fig. 3.2C). RDM was up to 214% higher for dianthus 'Amazon Rose Magic' young plants grown
under a moderate DLI compared to a low DLI. node nor leaf number, or off-type incidence was
influenced by photoperiod or DLI for either cultivar tested (data not reported).
Time to visible flower bud and node count
Young-plant and finishing photoperiod interacted to influence time to VB (TVB) for both
cultivars tested under both DLIs. However, this influence was weak and is unlikely to have a
commercial impact. For instance, dianthus ‘Amazon Neon Cherry’ finished under a 11-h
photoperiod and a moderate DLI initiated VBs only ≈2 d faster when grown under a 16-h young-
plant photoperiod compared to a 9-h young plant photoperiod. Similarly, plants finished under a
16-h photoperiod and a moderate DLI, and started under a 16-h photoperiod, initiated VBs
within 1 d of plants started under a 9-h photoperiod (Fig. 3.3A). Additionally, plants finished
under a 16-h photoperiod and a moderate DLI initiated VBs only 1 and 2 d faster than those
finished under 11-h photoperiods, when the young-plant photoperiod was 9- and 16-h,
respectively (Fig. 3.3A). Similar trends were observed for dianthus ‘Amazon Rose Magic’ (Fig.
3.4A). Plants of both cultivars finished under a low DLI demonstrated similar responses,
although delayed. (Fig. 3.3B; 3.3C; 3.4B; 3.4C). While photoperiod did not strongly influence
TVB, dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’ finished under a low DLI
took an average of 11 and 14 d longer, respectively, to reach VB than those finished under a
moderate DLI. Moreover, node number below VB was negligibly influenced by young-plant and
91
finishing photoperiods, as well as DLI. Plants of both cultivars had an average of 16 to 17 nodes
below their first VB regardless of treatment (data not reported).
Time to open flower
Young-plant and finishing photoperiod interacted to weakly influence time to OF (TOF)
of dianthus ‘Amazon Neon Cherry’, though TOF was negligibly influenced by young-plant
photoperiod. For example, plants reached OF ≈2 d faster when finished under a 16-h photoperiod
compared to an 11-h photoperiod (Fig. 3.5A). Moreover, young-plant and finishing photoperiods
independently influenced TOF of dianthus ‘Amazon Rose Magic’. Under a moderate DLI,
young-plant photoperiod did not commercially influence TOF (Fig. 3.6A). TOF was delayed by
≈3 d, or hastened by ≈2 d, when finished under a 16-h photoperiod compared to an 11-h
photoperiod, during reps. 1 and 2, respectively (Fig. 3.6B). TOF of both cultivars was delayed
when finished under a low DLI compared to a moderate DLI, although the general trends were
similar. (Fig. 3.5B; 3.5C; 3.6D). Dianthus cut flowers of both cultivars reached OF 11 d faster on
average when finished under a moderate DLI compared to a low DLI.
Time to harvest
Young-plant and finishing photoperiods interacted to influence time to harvest (TTH) of
dianthus ‘Amazon Neon Cherry’. However, plants finished under a moderate DLI were
negligibly influenced by young-plant photoperiods for both replications. TTH was 1 d slower
and 4 d faster for plants finished under a 16-h photoperiod compared to an 11-h photoperiod
during reps. 1 and 2, respectively (Fig. 3.7A; 3.7B). Young-plant and finishing photoperiods
acted independently to influence TTH of moderate-DLI-grown dianthus ‘Amazon Rose Magic’.
92
While young-plant photoperiod did not commercially influence TTH (Fig. 3.8A), TTH was up to
3 d slower and 2 d faster for plants finished under a 16-h photoperiod compared to an 11-h
photoperiod for reps. 1 and 2, respectively (Fig. 3.8B).
Cut flowers were reliably and consistently harvestable when finished under a moderate
DLI. However, only up to 57% and 32% of dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose
Magic’ plants finished under a low DLI were harvestable by the end of the study (13–15 weeks
after transplant), respectively. Conversely, ≈99% of cut flowers finished under a moderate DLI
became harvestable by the end of the study. Additionally, while neither young-plant or finishing
photoperiod statistically influenced TTH of dianthus ‘Amazon Rose Magic’ finished under a low
DLI, TTH of dianthus ‘Amazon Neon Cherry’ finished under a low DLI was 2 d faster when
finished under a 16-h photoperiod compared to an 11-h photoperiod. TTH of dianthus ‘Amazon
Neon Cherry’ and ‘Amazon Rose Magic’ grown under a low DLI was 9 and 10 d slower on
average, respectively, than those finished under a moderate DLI.
Cut flower morphology at harvest
Young-plant and finishing photoperiods interacted to influence dianthus ‘Amazon Neon
Cherry’ stem length at harvest, whereas stem length of dianthus ‘Amazon Rose Magic’ was only
influenced by finishing photoperiod. For instance, dianthus ‘Amazon Neon Cherry’ cut flowers
started under a 16-h photoperiod were ≈3 cm longer at harvest compared to those started under a
9-h photoperiod when both were finished under an 11-h photoperiod. Similarly, when finished
under a 16-h photoperiod, plants started under a 16-h photoperiod were up to 4 cm longer at
harvest compared to those started under a 9-h photoperiod (Fig. 3.9A; 3.9C). Moreover, plants
started under a 9-h young-plant photoperiod and finished under a 16-h photoperiod were up to
93
≈16 cm longer at harvest than those finished under an 11-h photoperiod. Furthermore, plants
started under a 16-h young plant photoperiod and finished under a 16-h photoperiod were up to
≈11 cm longer than those finished under an 11-h photoperiod (Fig. 3.9A; 3.9C). Dianthus
‘Amazon Rose Magic’ stem length increased from ≈69 to up to 82 cm as the finishing
photoperiod increased from 11- to 16-h (Fig. 3.10A), while young-plant photoperiod did not
significantly influence stem length at harvest. Plants finished under a low DLI exhibited a similar
response, and cut flowers were of similar length to those finished under moderate DLIs (Fig.
3.9B; 3.10B; 3.10C).
Stem caliper of dianthus ‘Amazon Rose Magic’ at harvest was not significantly and/or
commercially influenced by either young-plant or finishing photoperiod. Under a moderate DLI,
all cut flowers were ≈13 mm thick. Additionally, cut flowers finished under a low DLI were not
significantly influenced by photoperiod delivered during either stage, but were ≈2 mm thinner
than those finished under a moderate DLI (data not reported). Moreover, dianthus ‘Amazon
Neon Cherry’ and ‘Amazon Rose Magic’ cut flowers had up to 2 and 1 more branches at harvest,
respectively, when finished under an 11-h photoperiod compared to a 16-h photoperiod, while
young-plant photoperiod did not have a significant influence. Branch number was not influenced
by photoperiod when plants were finished under a low DLI, however, dianthus ‘Amazon Neon
Cherry’ and ‘Amazon Rose Magic’ plants had ≈8 and 7 fewer branches at harvest compared to
those finished under a moderate DLI (data not reported). Furthermore, young-plant or finishing
photoperiod did not commercially influence inflorescence number at harvest for either cultivar
tested. However, dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’ finished under low
DLIs had an average of 6 and 8 fewer inflorescences at harvest, respectively, than those finished
under moderate DLIs (data not reported).
94
Discussion
Overall, young-plant photoperiod had a negligible, although statistically significant,
influence on development time and finished quality of both dianthus cultivars tested. This was
inconsistent with our hypothesis, which was that young plants grown under longer photoperiods
would develop faster than those grown under shorter photoperiods during finishing. Moreover,
while finishing photoperiod had a marginal influence on TVB, TOF, and TTH compared to
young-plant photoperiod, it wasn’t strong enough to have a meaningful commercial impact. For
instance, dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’ finished under a 16-h
photoperiod became harvestable only 1-2 d earlier, on average, than those finished under an 11-h
photoperiod. Based on our findings, we characterize dianthus ‘Amazon Neon Cherry’ and
‘Amazon Rose Magic’ as day-neutral plants. This is consistent with Currey et al. (2011), who
reported that Dianthus barbatus is a day-neutral plant. However, dianthus plants generally
demonstrate a facultative LDP response to photoperiod. In fact, dianthus ‘Amazon’ cut flowers
are currently characterized as facultative LDPs (PanAmerican Seed, 2021).
Despite the minimal influence of young-plant and finishing photoperiod on development
time, DLI strongly influenced TVB, TOF, and TTH of both cultivars tested. For example, TVB
of dianthus finished under a low DLI was 11 to 14 d longer those finished under moderate DLIs.
Furthermore, TTH of dianthus finished under a low DLI was 9 to 10 d slower on average than
those finished under a moderate DLI. This is consistent with prior research on DLI, reporting
that celosia ‘Gloria Mix’ (Celosia argentea var. plumosa), and marigold ‘Bonanza Yellow’
(Tagetes patula) flowered 10 and 5 d earlier, respectively, when young plants were grown under
a DLI of 14.2 mol∙m–2∙d–1 compared to those grown under a DLI of 4.1 mol∙m–2∙d–1 (Pramuk and
Runkle, 2005). In a separate study, gladiolus ‘Dr. Fleming’ corms exposed to 80% shade for 5 d
95
yielded 23% less flowering plants compared to unshaded corms (Shillo and Halevy, 1976).
Moreover, Cavins and Dole (2001) found that campanula ‘Champion Blue’ (Campanula
medium) grown under SL providing ≈90 µmol·m–2·s–1 for 8 h·d–1 reached anthesis up to 14 d
faster compared to plants grown without SL.
Flower bud formation, flowering, and harvest were generally slightly hastened when
plants were finished under LDs. Contrary to our hypothesis, plants finished under a 16-h
photoperiod had longer stems than those finished under any other photoperiod. This is consistent
with Talon and Zeevaart (1992), who reported that after 20 d the LDP Sweet William catchfly
(Silene armeria) was up to 17 cm longer when grown under a constant 16-h photoperiod
compared to those grown under a constant 8-h photoperiod. Moreover, plants grown under an
8-h photoperiod before being placed under a 16-h photoperiod were only 12 cm longer than
those grown under a constant 8-h photoperiod (Talon and Zeevaart, 1992). This is further
supported by Cavins and Dole (2001), who reported that the LDPs campanula ‘Champion Blue’
and lupine ‘Bright Gems’ (Lupnius hartwegii) finished under a 16-h photoperiod were up to
16.9% and 23.2% longer at harvest compared to those finished under an 8-h photoperiod.
Conversely, campanula ‘Champion Pink’ was 26.9% longer at harvest when finished under an 8-
h photoperiod compared to a 16-h photoperiod. Furthermore, Heins and Wilkins (1977) reported
that dianthus ‘White Sim’ developed ≈8 more branches when grown under a 9-h SD compared to
inductive LDs created with a 5-h NI. This is consistent with the fact that both dianthus cultivars
tested in the present study had 1-2 more branches at harvest when finished under a 11-h SD
compared to a 16-h LD.
To our surprise, plants finished under a low DLI were comparable in length to those
finished under a moderate DLI (Fig. 3.9; 3.10). However, this is consistent with prior literature
96
indicating that finished oriental lily ‘Star Gazer’ (Lilium spp.) cut flowers grown under 60
µmol∙m–2∙s–1 of SL provided by HPS lamps for 10 h∙d–1 were only 8% shorter, on average, than
those grown without SL (Treder and Kubik, 2000). Moreover, Cavins and Dole (2001) reported
that campanula ‘Champion Blue’ and ‘Champion Pink’ were up to 7.3% and 18.3% longer when
finished without SL compared to plants finished under SL providing ≈90 µmol·m– 2·s–1 for 8 h·d–
1
.
Dianthus finished under a moderate DLI had 6-8 more inflorescences at harvest, on
average, than those finished under a low DLI. It is well documented that higher DLIs typically
produce plants with more flowers than lower DLIs. For instance, Faust et al. (2005) reported that
Begonia ‘Vodka Cocktail’ (Begonia ×semperflorens-cultorum), impatiens ‘Cajun Red’
(Impatiens walleriana), petunia ‘Apple Blossom’ (Petunia ×hybrida), vinca ‘Pacific Lilac’
(Catharanthus roseus), and zinnia ‘Dreamland Rose’ (Zinnia elegans) produced ≈9, 3, 3, 2, and
1 more flowers under a DLI of 19 mol∙m–2∙d–1 compared to 5 mol∙m–2∙d–1. Oh et al. (2009) found
that cyclamen ‘Cultivar’ (Cyclamen persicum) grown under a DLI of 11.5 mol∙m–2∙d–1 had ≈25
more branches after 16 weeks than those grown under a DLI of 1.4 mol∙m–2∙d–1. The number of
lateral inflorescences of yarrow ‘Red Velvet’ (Achillea millefolium) and beeblossom ‘Whirling
Butterflies’ (Gaura lindheimeri) increased by 171 and ≥60%, respectively, as the DLI increased
from 5 to 20 mol∙m–2∙d–1 (Fausey et al., 2005).
Moreover, dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’ finished under
moderate DLIs developed stems that were up to 17.4% and 29.6% thicker, respectively, than
those finished under low DLIs. Contrary to this, campanula ‘Champion Blue’ and ‘Champion
Pink’ were up to 12.5% and 5.8% thicker when finished without SL compared to plants finished
under SL providing ≈90 µmol·m–2·s–1 for 8 h·d–1 (Cavins and Dole, 2001).
97
Only up to 57 and 32% of dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’
cut flowers were harvestable, respectively, when finished under a low DLI. Conversely, a
moderate DLI yielded up to 99% and 98% of planted cut flowers, respectively. A significant
portion of dianthus ‘Amazon Rose Magic’ cut flowers finished under a low DLI exhibited a
wider, partially-full inflorescence compared to those finished under a moderate DLI (Fig. 3.11).
This phenotype was also observed with dianthus ‘Amazon Neon Cherry’ cut flowers, although
less frequently. Such cut flowers were deemed unharvestable because they did not meet market
specifications. To our knowledge, this phenotype has not yet been reported in literature.
Considering the weak developmental and physiological responses exhibited by young
plants and finished cut flowers grown under different young-plant photoperiods, we advise that
dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’ young plants be grown under any
photoperiod between 9- and 16-h for four weeks, with a moderate DLI (e.g., ≥10 mol∙m–2∙d–1) to
promote biomass accumulation. Additionally, plants can be finished under any photoperiod
between 11 and 16 h, or a 4-h NI, for timely harvests of high-quality cut flowers. However,
plants should be finished under a 16-h photoperiod if longer stems are desired. Additionally,
plants should be finished under a moderate DLI (e.g., ≥14 mol∙m–2∙d–1) to decrease production
time while ensuring sturdy, marketable cut flowers.
Acknowledgements
We gratefully acknowledge BloomStudios and The Association of Specialty Cut Flower
Growers for providing funding and supplies. We also gratefully acknowledge Nate DuRussel,
John Gove, and Ian Holcomb for greenhouse assistance, Hydrofarm for netting, Ludvig
Svensson for shade cloth, Raker Roberta’s Young Plants for sowing seeds, Syndicate Sales for
98
floral supplies, and Signify for LED supplemental lighting fixtures. This work was supported by
the USDA National Institute of Food and Agriculture, Hatch project MICL02472.
99
APPENDIX
100
Table 3.1. Actual daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], air average daily temperatures (ADTs), mean day
temperatures, and mean night temperatures [mean ± SD (°C)] throughout the duration of the young-plant stage for reps. 1 and 2.
Photoperiod Moderate DLI Low DLI
(h)
DLI ADT Day Night DLI ADT Day Night
[mean ± SD [mean ± SD [mean ± SD [mean ± SD [mean ± SD [mean ± [mean ± SD [mean ±
(mol·m–2·d–1)] (°C)] (°C)] (°C)] (mol·m–2·d–1)] SD (°C)] (°C)] SD (°C)]
Rep. 1
9 9.7 ± 1.6 20.6 ± 1.2 23.2 ± 2.4 19.5 ± 2.1 6.5 ± 1.8 20.2 ± 0.5 22.5 ± 1.8 19.3 ± 1.3
10 8.4 ± 2.8 20.5 ± 0.5 22.6 ± 2.2 19.7 ± 1.2 6.1 ± 1.9 19.7 ± 0.8 22.1 ± 2.0 18.8 ± 1.9
z
11 19.4 ± 0.5 21.5 ± 2.1 18.5 ± 1.3 6.0 ± 2.5 20.2 ± 0.5 22.6 ± 2.0 19.3 ± 1.3
12 9.8 ± 2.0 18.7 ± 2.0 21.6 ± 3.0 17.1 ± 3.5 6.8 ± 2.2 19.9 ± 0.6 22.3 ± 2.1 19.0 ± 1.6
13 9.2 ± 3.1 20.1 ± 0.7 22.5 ± 2.0 19.0 ± 1.6 6.6 ± 2.3 19.7 ± 0.8 22.3 ± 2.3 18.5 ± 2.0
z
15 9.8 ± 2.3 19.8 ± 0.7 21.8 ± 1.9 19.1 ± 1.5 19.5 ± 1.2 22.3 ± 2.5 18.2 ± 2.5
16 8.6 ± 2.6 20.6 ± 1.2 22.7 ± 1.7 19.9 ± 2.1 6.4 ± 2.1 20.0 ± 0.6 22.5 ± 2.4 19.0 ± 1.7
Rep. 2
9 7.4 ± 1.3 19.9 ± 1.1 21.7 ± 2.7 18.0 ± 2.5 5.9 ± 2.0 19.3 ± 0.48 20.9 ± 1.9 17.7 ± 1.35
10 9.6 ± 1.4 18.3 ± 1.3 20.7 ± 2.1 15.8 ± 2.7 5.7 ± 1.6 18.9 ± 0.49 21.1 ± 2.7 16.5 ± 1.3
11 9.5 ± 1.2 20.1 ± 1.3 22.3 ± 2.8 19.1 ± 2.8 6.5 ± 1.6 18.5 ± 0.6 20.7 ± 2.3 16.4 ± 1.5
12 9.3 ± 1.7 20.0 ± 0.4 21.7 ± 2.6 18.3 ± 1.9 6.6 ± 1.8 19.3 ± 0.7 21.0 ± 1.8 17.6 ± 1.5
13 9.6 ± 1.3 20.7 ± 0.9 22.0 ± 2.2 19.4 ± 2.3 6.0 ± 1.7 19.1 ± 1.5 21.0 ± 2.2 17.1 ± 2.9
15 10.0 ± 2.8 20.6 ± 1.2 21.7 ± 2.2 19.4 ± 2.6 7.5 ± 2.4 18.7 ± 0.5 21.1 ± 2.7 16.5 ± 1.25
16 10.2 ± 1.9 19.4 ± 0.4 21.1 ± 2.0 17.6 ± 1.3 6.9 ± 2.3 18.7 ± 0.7 20.8 ± 2.7 16.6 ± 1.5
z
No data recorded
101
Table 3.2. Actual daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], air average daily temperatures (ADTs), mean day
temperatures, and mean night temperatures [mean ± SD (°C)] throughout the duration of the finishing stage for reps. 1 and 2.
Photoperiod Moderate DLI Low DLI
(h)
DLI ADT Day Night DLI ADT Day Night
[mean ± SD [mean ± SD [mean ± SD [mean ± [mean ± SD [mean ± [mean ± [mean ±
(mol·m–2·d–1)] (°C)] (°C)] SD (°C)] (mol·m–2·d–1)] SD (°C)] SD (°C)] SD (°C)]
Rep. 1
11 13.4 ± 3.9 20.9 ± 1.6 22.5 ± 3.2 19.3 ± 3.1 4.8 ± 2.1 19.8 ± 1.4 21.3 ± 2.8 18.4 ± 2.9
12 13.6 ± 4.2 20.6 ± 1.1 22.4 ± 3.2 18.8 ± 2.8 –z 19.8 ± 1.7 21.3 ± 3.0 18.2 ± 3.5
13 14.5 ± 4.9 19.0 ± 1.6 21.3 ± 3.8 16.7 ± 3.0 5.4 ± 2.3 20.0 ± 1.3 21.4 ± 2.9 18.5 ± 2.8
14 –z 20.2 ± 1.6 21.7 ± 3.4 18.7 ± 3.0 5.7 ± 2.5 19.9 ± 1.4 21.5 ± 3.1 18.2 ± 3.1
15 12.8 ± 3.4 20.0 ± 1.2 21.5 ± 3.0 18.5 ± 2.6 5.1 ± 2.0 19.9 ± 1.3 21.3 ± 2.8 18.5 ± 2.7
16 13.2 ± 4.9 20.3 ± 1.6 22.1 ± 3.7 18.5 ± 3.5 4.9 ± 2.4 19.8 ± 1.2 21.3 ± 2.6 18.4 ± 2.5
4-h NI 13.9 ± 4.5 20.9 ± 1.4 22.5 ± 3.3 19.3 ± 2.5 4.9 ± 2.4 19.8 ± 1.2 21.3 ± 2.6 18.4 ± 2.5
Rep. 2
11 13.7 ± 5.0 20.7 ± 1.4 23.7 ± 3.2 17.7 ± 2.2 5.4 ± 2.0 19.6 ± 1.2 21.9 ± 3.5 17.1 ± 2.6
12 13.7 ± 5.3 20.5 ± 1.6 23.4 ± 3.4 17.5 ± 2.5 5.8 ± 2.0 19.6 ± 1.2 21.7 ± 3.3 17.4 ± 2.6
13 13.0 ± 4.3 19.7 ± 1.9 22.3 ± 3.2 16.9 ± 2.2 5.5 ± 2.0 19.1 ± 1.6 21.2 ± 3.2 16.7 ± 3.2
14 14.6 ± 4.8 17.9 ± 1.5 21.4 ± 3.4 14.4 ± 2.0 4.7 ± 2.2 19.7 ± 1.2 21.5 ± 2.7 17.9 ± 2.2
15 14.5 ± 4.8 20.1 ± 1.4 22.9 ± 3.1 17.3 ± 1.9 5.6 ± 2.3 19.2 ± 1.4 20.9 ± 3.0 16.9 ± 2.2
16 13.7 ± 4.5 20.9 ± 1.4 23.9 ± 3.2 17.8 ± 2.1 5.5 ± 2.1 19.3 ± 1.3 21.3 ± 3.1 17.1 ± 2.6
4-h NI 14.0 ± 8.3 20.6 ± 1.6 23.6 ± 3.4 17.6 ± 2.3 5.9 ± 2.4 19.6 ± 1.2 21.4 ± 2.7 17.8 ± 2.2
z
No data recorded
102
Table 3.3. Regression analysis equations and adjusted R2 for height; root dry mass; and shoot dry
mass in response to photoperiod (P; 9-, 11-, 12-, 13-, 14-, 15-, 16-, 18-, 24-h photoperiods or a 4-
h NI) of dianthus ‘Amazon Neon Cherry’ and ‘Amazon Rose Magic’ (Dianthus barbatus
interspecific) young plants grown under a moderate or very low DLI. All models are in the form
of: ƒ = y0 + a*P + b*P2.
Parameter y0 a b R2
Dianthus ‘Amazon Neon Cherry’
Moderate DLI
Height (cm) -1.83z 1.01 -0.03 0.854
Root dry mass (g)
Rep. 1 -0.01 0.00 -8.35E-05 0.353
Rep. 2 0.01 0.00 6.93E-05 0.356
Shoot dry mass (g)
Rep. 1 0.00 0.01 0.00 0.294
Low DLI
Height (cm) 2.05 0.36 -0.01 0.651
Shoot dry mass (g) 0.03 0.00 5.17E-05 0.217
Dianthus ‘Amazon Rose Magic’
Moderate DLI
Height (cm) -0.68z 0.87 -0.27 0.541
Root dry mass (g) -0.02 0.00 0.00 0.390
Shoot dry mass (g)
Rep. 1 0.00 0.01 0.00 0.166
Low DLI
Height (cm) 2.15 0.35 0.00 0.758
Root dry mass (g)
Rep. 2 0.01 0.00 1.47E-05 0.142
Shoot dry mass (g)
Rep. 2 0.06 -0.01 0.00 0.428
z
Coefficients for model equations were used to generate Figs. 3.1A-E; 3.2A-F.
103
Table 3.4. Regression analysis equations and adjusted R2 for time to visible flower bud, time to open flower, time to harvest, and stem
length at harvest in response to young-plant photoperiod (YP; 9, 11, 12, 13, 15, or 16 h) and/or finishing photoperiod (FP; 11, 12, 13,
14, 15, or 16 h, or a 4-h NI) of dianthus ‘Amazon Neon Cherry’ (Dianthus barbatus interspecific). All models are in the form of: ƒ =
a*YP + b*FP + c*YP2 + d*FP2+e*(YP*FP) unless otherwise indicated.
Parameter y0 a b c d e R2
Time to visible flower bud (d)
Moderate DLI y
0.85z 8.30 0.04 -0.27 -0.12 0.264
Rep.1 Low DLI 1.13 10.30 -0.01 -0.39 -0.06 0.302
Rep. 2 Low DLI 2.50 7.20 -0.01 -0.20 -0.15 0.240
Time to open flower (d)
Moderate DLI 1.49 9.73 0.02 -0.32 -0.14 0.233
Rep.1 Low DLI 0.84 12.52 0.01 -0.46 -0.08 0.199
Rep. 2 Low DLI 1.92 9.80 0.00 -0.31 -0.13 0.122
Time to harvest (d)
Rep. 1 Moderate DLI 129.15 -0.94 -7.54 0.04 0.27 0.270
Rep. 2 Moderate DLI -0.59 12.49 0.06 -0.46 -0.07 0.517
Stem length at harvest (cm)
Rep. 1 Moderate DLI 7.27 2.99 -0.20 0.03 -0.15 0.597
Rep. 2 Moderate DLI 2.35 8.61 -0.03 -0.22 -0.13 0.310
Rep. 1 Low DLI* 70.91 0.20 0.03 0.323
Rep. 2 Low DLI* 217.21 -21.98 0.86 0.432
z
Coefficients for model equations were used to generate Figs. 3.3, 3.5, 3.7, and 3.8.
y
Blank cells = 0
* ƒ = y0 + a*P + b*P2; P = Finishing photoperiod
104
Table 3.5. Regression analysis equations and adjusted R2 for time to visible flower bud, time to
open flower, time to harvest, and stem length at harvest in response to young-plant photoperiod
(YP; 9, 11, 12, 13, 15, or 16 h) and/or finishing photoperiod (FP; 11, 12, 13, 14, 15, or 16 h, or a
4-h NI) of dianthus ‘Amazon Rose Magic’ (Dianthus barbatus interspecific). All models are in
the form of: ƒ = a*YP + b*FP + c*YP2 + d*FP2+e*(YP*FP) unless otherwise indicated.
Parameter y0 a b c d e R2
Time to visible flower bud (d)
Moderate DLI y
3.40z 6.02 -0.09 -0.20 -0.09 0.101
Rep.1 Low DLI -1.56 12.72 0.13 -0.43 -0.14 0.287
Rep. 2 Low DLI 4.14 6.56 -0.06 -0.16 -0.20 0.111
Time to open flower (d)
Rep. 1 Moderate DLI* 110.37 -5.89 0.22 0.156
Rep. 2 Moderate DLI* 57.81 2.80 -0.14 0.204
Pooled moderate DLI** 66.52 0.63 -0.02 0.039
Rep. 1 Low DLI* 63.81 3.60 -0.16 0.276
Rep. 2 Low DLI* 94.11 -1.98 0.06 0.172
Rep. 1 Low DLI** 108.52 -3.95 0.15 0.206
Rep. 2 Low DLI** 71.75 1.20 -0.05 0.114
Time to harvest (d)
Rep. 1 Moderate DLI* 117.13 -6.49 0.24 0.150
Rep. 2 Moderate DLI* 58.95 2.88 -0.14 0.355
Pooled moderate DLI** 69.69 0.44 -0.01 0.062
Stem length at harvest (cm)
Rep. 1 Moderate DLI* 116.04 -8.54 0.40 0.597
Rep. 2 Moderate DLI* 30.94 4.80 -0.12 0.513
Rep. 1 Low DLI 4.09 6.13 -0.09 -0.10 -0.16 0.329
Rep. 2 Low DLI 232.79 -2.30 -23.34 0.09 0.92 0.527
z
Coefficients for model equations were used to generate Figs. 3.4, 3.6, 3.8, and 3.10.
y
Blank cells = 0
* ƒ = y0 + a*P + b*P2; P = Finishing photoperiod
** ƒ = y0 + a*P + b*P2; P = Young-plant photoperiod
105
Figure 3.1. Effect of 9-, 10-, 11-, 12-, 13-, 15-, and 16-h young-plant photoperiods on the height
(A; D), shoot dry mass (B; E), and root dry mass (C; F) of dianthus ‘Amazon Neon Cherry’
(Dianthus barbatus interspecific) young plant growth responses. Black symbols indicate means;
error bars indicate standard error of the mean. Coefficients are presented in Table 3.3.
106
Figure 3.2. Effect of 9-, 10-, 11-, 12-, 13-, 15-, and 16-h young-plant photoperiods on the height
(A; D), shoot dry mass (B; E), and root dry mass (C; F) of dianthus ‘Amazon Rose Magic’
(Dianthus barbatus interspecific) young plant growth responses. Black symbols indicate means;
error bars indicate standard error of the mean. Coefficients are presented in Table 3.3.
107
A B
80
80
(d)
bud (d)
Time to visible bud Time to visible
70
70
60 60 )
(h (h
)
pe pe
rio rio
16 d
50 16 d 50 to
to 14
14 ho
ho 12 tp
12 tp NI
18
18
NI
NI 16 pl
16 10
-p 10 an
14 la 14
Finish n Finish 12 g-
in 12 ing photop
g pho ng un
toperi Yo eriod Yo
od (h) u (h)
C
80
bud (d)
Time to visible
70
60 )
(h
eriod
16
50 op
14 ho t
12 tp
18NI
16 10 pl
14 an
Finish 12 ng
ing photop -
eriod Yo
u
(h)
Figure 3.3. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to visible flower bud of dianthus
‘Amazon Neon Cherry’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-
grown cut flowers, (B) low-DLI-grown cut flowers from replication 1, and (C) low-DLI-grown
cut flowers from replication 2. Black symbols represent individual data points for sequential
photoperiods; red symbols represent means from NI treatments. NI means were excluded from
response surfaces. Model predictions are represented by response surfaces; coefficients are
presented in Table 3.4.
108
A B
80 80
bud (d)
75 (d) 75
bud
Time to visible Time to visible
70 70
65 65
(h) (h)
60 60
pe pe
rio rio
55 16 d 55 16 d
14 to 14 to
50 tp 50 tp
12 ho 12 ho
18
NI NI
18
16 10 -p 16 10-p
14 lan 14 lan
Finish 12 Finish 12
in g pho un g in g pho un g
toper toper
iod (h Yo iod (h Yo
) )
C
80
(d) 75
bud
Time to visible
70
65
)
(h
60
pe
rio
55 16 d
14 to
50
ho
12 tp
18
NI
16 10 lan
14
Finish 12
-p
in g pho Yo
toper un
iod (h g
)
Figure 3.4. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to visible flower bud of dianthus
‘Amazon Rose Magic’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-
grown cut flowers, (B) low-DLI-grown cut flowers from replication 1, and (C) low-DLI-grown
cut flowers from replication 2. Black symbols represent individual data points for sequential
photoperiods; red symbols represent means from NI treatments. NI means were excluded from
response surfaces. Model predictions are represented by response surfaces; coefficients are
presented in Table 3.5.
109
A B
wer (d)
90 90
(d)
Time to open flo
wer
Time to open flo
80 80
(h) (h)
70 pe 70 pe
rio rio
16 d 16 d
14 to 14 to
60 pl
12an 60 pl
12an
18NI tp 18NI tp
16 10 ho 16 10 ho
Finish 14 Finish 14
in 12 ng ing p 12 ng
g pho - ho toper -
toper Yo iod (h Yo
iod (h u ) u
)
C
wer (d)
90
Time to open flo
80
)
(h
70 eriod
16
op
14 ho
60 t
12 tp
NI
18
16 ng
10
Finish 14 -p
in 12 lan
g pho
toper Yo
iod (h u
)
Figure 3.5. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to open flower of dianthus ‘Amazon
Neon Cherry’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-grown cut
flowers, (B) low-DLI-grown cut flowers from replication 1, and (C) low-DLI-grown cut flowers
from replication 2. Black symbols represent individual data points for sequential photoperiods;
red symbols represent means from NI treatments. NI means were excluded from response
surfaces. Model predictions are represented by response surfaces; coefficients are presented in
Table 3.4.
110
90
A B
Time to open flower (d)
85
80
75
70
65
0
90 C D 12 14 16 18
Time to open flower (d)
Finishing photoperiod (h)
85
80
75
70 Rep. 1 Rep. 1
Rep. 2 Rep. 2
65 Rep. 1 means Rep. 1 means
Rep. 2 means Rep. 2 means
0
10 12 14 16 12 14 16 NI
18
Young-plant photoperiod (h) Finishing photoperiod (h)
Figure 3.6. Individual effects of young-plant (9, 11, 12, 13, 15, or 16 h) or finishing photoperiod
(11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to open flower of dianthus ‘Amazon Rose Magic’
(Dianthus barbatus interspecific). Figures represent (A and B) moderate-DLI-grown cut flowers
and (C and D) low-DLI-grown cut flowers. Black symbols indicate means; error bars indicate
standard error of the mean. NI means were excluded from regressions. Coefficients are presented
in Table 3.5.
111
Figure 3.7. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and/or finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to harvest of dianthus ‘Amazon Neon
Cherry’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-grown cut
flowers from replication 1, and (B) moderate-DLI-grown cut flowers from replication 2. Black
symbols represent individual data points for sequential photoperiods; red symbols represent
means from NI treatments. NI means were excluded from response surfaces. Model predictions
are represented by response surfaces; coefficients are presented in Tables 3.4.
112
80
78 A B
Time to harvest (d)
76
74
72
70
68 Rep. 1
Rep. 2
66 Rep. 1 means
Rep. 2 means
0
10 12 14 16 12 14 16 NI
18
Young-plant photoperiod (h) Finishing photoperiod (h)
Figure 3.8. Individual effects of young-plant (9, 11, 12, 13, 15, or 16 h) or finishing photoperiod
(11, 12, 13, 14, 15, 16 h, or a 4-h NI) on time to harvest of dianthus ‘Amazon Rose Magic’
(Dianthus barbatus interspecific) finished under a moderate DLI. Black symbols indicate means;
error bars indicate standard error of the mean. Coefficients are presented in Table 3.5.
113
A B
95
Stem length at harvest (cm)
90
95
rvest (cm)
85
90 80
85 75
ha
Stem length at
80 70
(h) 65
75
an 60
70 tp 16
ho
14to
65 12 pe
18
NI rio 0
16 10 d
14 12 14 16 NI
18
Finish 12 un
ing ph g-p
otope l Finishing photoperiod (h)
riod ( Yo
h)
Rep. 1
Rep. 2
Rep. 1 means
Rep. 2 means
C
95
rvest (cm)
90
85
ha
Stem length at
80
(h)
75
an
70 tp 16
ho
14to
65 12 pe
18
NI rio
16 10 d
Finish 14 un
ing ph 12 g-p
otope l
riod (
h) Yo
Figure 3.9. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and/or finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on stem length at harvest of dianthus ‘Amazon
Neon Cherry’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-grown cut
flowers from replication 1, (B) low-DLI-grown cut flowers from replication 1 and 2, and (C)
moderate-DLI-grown cut flowers from replication 2. Black symbols in Fig. 9A and C represent
individual data points for sequential photoperiods; red symbols represent means from NI
treatments. Black symbols in Fig. 3.9B indicate means; error bars indicate standard error of the
mean. NI means were excluded from regressions. Model predictions are represented by response
surfaces or regressions; coefficients are presented in Table 3.4.
114
A B
85
Stem length at harvest (cm)
80 90
harvest (cm)
85
75
80
Stem length at
70 75
)
(h
70
65
ph
ot
16
65 op
14 er
60
iod
12
0 NI
18
16 10 plan
12 14 16 NI
18 14 t
Finish 12 un
in g pho g-
Finishing photoperiod (h) toper
iod (h Yo
)
Rep. 1
Rep. 2
Rep. 1 means
Rep. 2 means
C
90
harvest (cm)
85
80
Stem length at
75
)
(h
70
pe
rio
65 16 d
14 to
60
ho
12 tp
18
NI
16 10 pl
14 an
Finish 12 g-
in g pho un
toper
iod (h Yo
)
Figure 3.10. Effects of young-plant photoperiod (9, 11, 12, 13, 15, or 16 h) and/or finishing
photoperiod (11, 12, 13, 14, 15, 16 h, or a 4-h NI) on stem length at harvest of dianthus ‘Amazon
Rose Magic’ (Dianthus barbatus interspecific). Figures represent (A) moderate-DLI-grown cut
flowers, (B) low-DLI-grown cut flowers from replication 1, and (C) low-DLI-grown cut flowers
from replication 2. Black symbols in Fig. 3.10A indicate means; error bars indicate standard
error of the mean. Black symbols in Figs. 3.10B and C represent individual data points for
sequential photoperiods; red symbols represent means from NI treatments. NI means were
excluded from regressions. Model predictions are represented by response surfaces or
regressions; coefficients are presented in Table 3.5.
115
Figure 3.11. Observational differences between dianthus ‘Amazon Rose Magic’ (Dianthus
barbatus interspecific) cut flowers finished under a moderate DLI or a low DLI.
116
REFERENCES
117
REFERENCES
Ahmad, I., Rafiq, M.B., Dole, J.M., Abdullah, B., Habib, K., 2017. Production and postharvest
evaluation of selected exotic specialty annual cut flower species in Punjab, Pakistan.
HortTechnology 27, 878–883.
Blacquière, T., Krijger, D., de Graaf-van der Zande, M., 2002. Photoperiodic lighting of
matricaria (Tanacetum parthenium). Acta Hortic. 580, 117–126.
Carman, H., 2007. California continues to lead nation in nursery and floral production. CORF
News 11, 1–4.
Cavins, T.J., Dole, J.M., 2001. Photoperiod, juvenility, and high intensity lighting affect
flowering and cut stem qualities of Campanula and Lupinus. HortScience 36, 1192–1196.
Currey, C.J., Lopez, R.G., Mattson, N.S., 2011. Flower induction of annuals. Purdue Extension
Bulletin H0-249-W.
(accessed 06 July 2022).
Dole, J.M., 2015. Photoperiod responses of ten specialty cut flowers. Acta Hortic. 1097, 163–
168.
Erwin, J.E., Warner, R.M., 2002. Determining the photoperiodic response group and effect of
supplemental irradiance on flowering of several bedding plant species. Acta Hortic. 580,
95–100.
Fausey, B.A., Heins, R.D., Cameron, A.C., 2005. Daily light integral affects flowering and
quality of greenhouse-grown Achillea, Gaura, and Lavandula. HortScience 40, 114–118.
Faust, J.E., Dole, J.M., 2021. Cut flowers and foliages. CAB International, Oxfordshire, United
Kingdom.
Faust, J.E., Holcombe, V., Rajapakse, N.C., Layne, D.R., 2005. The effect of daily light integral
on bedding plant growth and flowering. HortScience 40, 645–649.
Heins, R.D., Wilkins, H.F., 1977. Influence of photoperiod on improved ‘White Sim’ carnation
(Dianthus caryophyllus L.) branching and flowering. Acta Hortic. 71, 69–74.
Korczynski, P.C., Logan, J., Faust, J.E., 2002. Mapping monthly distribution of daily light
integrals across the contiguous United States. HortTechnology 12, 12–16.
Llewellyn, D., Schiestel, K., Zheng, Y., 2020. Increasing levels of supplemental LED light
enhances the rate flower development of greenhouse-grown cut gerbera but does not
affect flower size and quality. Agronomy 10, 1332.
118
Lopez, R.G., Runkle, E.S., 2008. Photosynthetic daily light integral during propagation
influences rooting and growth of cuttings and subsequent development of new guinea
impatiens and petunia. HortScience 43, 2052–2059.
Lugasi-Ben-Hamo, M., Kitron, M., Buston, A., Zaccai, M., 2010. Effect of shade regime on
flower development, yield and quality in lisianthus. Scientia Hortic. 124, 248–253.
Marcelis, L.F.M., Broekhuijsen, A.G.M., Meinen, E., Nijs, E.M.F.M., Raaphorst, M.G.M., 2006.
Quantification of the growth response to light quantity of greenhouse grown crops. Acta
Hortic. 711, 97–104.
Oh, W., Cheon, I.H., Kim, K.S., Runkle, E.S., 2009. Photosynthetic daily light integral
influences flowering time and crop characteristics of Cyclamen persicum. HortScience
44, 341–344.
PanAmerican Seed. 2021. GrowerFacts Dianthus 'Amazon™ F1 Series'.
Pramuk, L.A., Runkle, E.S., 2005. Photosynthetic daily light integral during seedling stage
influences subsequent growth and flowering of Celosia, Impatiens, Salvia, Tagetes, and
Viola. HortScience 40, 1336–1339.
Produce Marketing Association and Food Marketing Institute (PMAFMI), 2016. Trends in mass-
market floral. Produce Marketing Association, Newark, DE.
Runkle, E.S., 2008. LIP It! Greenhouse Prod. News 18, 66.
Runkle, E.S., 2013. Using the PGRs Collate and Florel. Greenhouse Prod. News 23, 58.
Runkle, E.S., Both, A.J., 2017. Delivering long-day lighting - technology options and costs. In:
Lopez, R.G., Runkle, E.S. (Eds.). Light management in controlled environments. Meister
Media Worldwide, Willoughby, OH, pp. 91–99.
Shillo, R., Halevy, A.H., 1976. The effect of various environmental factors on flowering of
gladiolus. I. Light intensity. Scientia Hortic. 4, 131–137.
Styer, R.C., 2002. Using Florel effectively. Greenhouse Prod. News 8, 10–15.
Talon, M., Zeevaart, J.A.D., 1992. Stem elongation and changes in the levels of gibberellins in
shoot tips induced by differential photoperiodic treatments in the long-day plant Silene
armeria. Planta 188, 457–461.
Treder, J., Kubik, M., 2000. Growth, flowering and partitioning of 14C-assimilate in oriental lily
‘Star Gazer’ as affected by light level and flower removal. J. of Hortic. Sci. and
Biotechnol. 75, 277–282.
119
U.S. Department of Agriculture (USDA), 2019. Floriculture crops 2018 summary.
(accessed 15 July 2020).
120
SECTION 4
121
Supplemental lighting quality influences time to flower and finished quality of three long-day
specialty cut flowers
Caleb E. Spall and Roberto G. Lopez*
Department of Horticulture, 1066 Bogue Street, Michigan State University, East Lansing, MI
48824-1325, USA
*Corresponding author. Tel.: +1 517-353-0342. E-mail address: rglopez@msu.edu (R.G. Lopez).
122
Abstract
Year-round demand for locally sourced specialty cut flowers continues to increase.
However, growers in northern latitudes cannot produce cut flowers outdoors or in high-tunnels
year-round due to low radiation intensities and temperatures. As a result, growers must utilize
greenhouses, but limited production information detailing manipulation of the radiation
environment exists. Therefore, our objective was to quantify the influence of supplemental
lighting (SL) quality on time to flower and harvest and stem quality of three long-day specialty
cut flowers. Godetia ‘Grace Rose Pink’ (Clarkia amoena), snapdragon ‘Potomac Royal’
(Antirrhinum majus), and stock ‘Iron Rose’ (Matthiola incana) plugs were transplanted into bulb
crates and placed in one of six greenhouse compartments with SL providing a total photon flux
density of 120 µmol·m–2·s–1 from 0700 to 1900 HR. After four weeks, SL was extended to
provide a 16-h photoperiod to induce flowering. SL treatments were provided by either high-
pressure sodium (HPS) fixtures or various light-emitting diode (LED) fixtures. Treatments were
defined by their 100-nm wavebands of blue (B; 400‒500 nm), green (G; 500‒600 nm), red (R;
600‒700 nm), and far-red (FR; 700‒800 nm) radiation (photon flux density in μmol·m–2·s–1) as
B7G60R44FR9 (HPS120), B20G50R45FR5, B20R85FR15, B30G25R65, B120, or R120. Time to harvest
(TTH) was up to 14, 15, and 10 d slower under R120 SL for godetia, snapdragon, and stock,
respectively, compared to the quickest treatments (HPS120, B120, and B20R85FR15 SL). However,
R120 SL produced cut flowers up to 18% longer than those grown under the quickest treatments.
Both broad-spectrum LED fixtures slightly delayed TTH compared to the quickest treatments.
Stem caliper was not commercially different between treatments for godetia or snapdragon,
although stems were up to 14% thinner for stock grown under B120 SL compared to the other
treatments. Flower petal color was not commercially different between SL treatments. We
123
recommend utilizing a SL fixture providing a spectrum similar to B20R85FR15 SL or either broad-
spectrum LED fixture, as they elicited desirable crop responses with minimal developmental,
quality, and visibility tradeoffs. While HPS lamps performed similarly to the recommended
fixtures, we recommend utilizing LEDs for their higher photon efficacy and potential energy
savings.
Keywords: high-pressure sodium lamps, light-emitting diodes, light quality, controlled-
environment agriculture, greenhouse
Abbreviations: ADT, average daily temperature; B, blue; DE, day extension; DLI, daily light
integral; FR, far red; G, green; LED, light-emitting diode; HPS, high-pressure sodium; LDP,
long-day plant; OF, open flower; PAR, photosynthetically active radiation; PFD, photon flux
density; PPFD, photosynthetic photon flux density; R, red; SL, supplemental lighting; TOF, time
to open flower; TTH, time to harvest; TVB, time to visible flower bud; VB, visible flower bud;
W, white.
124
Introduction
In the northern United States, specialty cut flower production can occur outdoors or in
unheated high tunnels when weather permits (Owen et al., 2016; Ortiz et al., 2012). However,
demand for locally-produced specialty cut flowers persists year round (Prinzing, 2021;
PMAFMI, 2016). During the winter and early spring, the outdoor solar daily light integral (DLI)
can fall to as low as 5 to 10 mol∙m–2∙d–1 (Faust and Logan, 2018), and as low as ≤5 mol∙m–2∙d–1
in controlled-environment greenhouses due to reflection from glazing and shading from the
superstructure (Lopez and Runkle, 2008). Additionally, low temperatures inhibit the production
of specialty cut flowers outdoors or in unheated high tunnels during these seasons. Because of
these unfavorable environmental conditions, controlled-environment greenhouses with high-
intensity supplemental lighting (SL) must be employed in these seasons to maintain
environmental conditions suitable for cut flower growth, so growers can tap into local markets
and satisfy consumer demand throughout the year.
In recent years, the advent of horticultural light-emitting diodes (LEDs) has given
growers the potential to further customize the emission spectra of their lighting sources, allowing
for the inclusion of narrow wavebands (Paradiso and Proietti, 2021; Kusuma et al., 2020; Owen
et al., 2018; Poel and Runkle, 2017). Thus, a large variety of SL fixtures with different static or
customizable emission spectra have become commercially available. The composition of
radiation emitted from a lighting fixture can have substantial effects on plant growth and
development, especially when the solar DLI is low (Runkle, 2019; Hernández and Kubota,
2017), with some wavebands acting not only as photosynthetic stimuli, but as developmental
signals (Goins et al., 1997). Photosynthetically active radiation (PAR; 400‒700 nm) is primarily
responsible for driving photosynthesis, although isolated wavebands within and outside of this
125
range can bring about specific photomorphogenic responses. Although outside of the traditional
definition of PAR, far-red (FR) radiation (700‒800 nm) has recently been shown to contribute to
photosynthesis directly by working synergistically with photons within the traditional
designation of PAR, and indirectly by promoting leaf expansion (Zhen et al., 2021; Zhen and
Bugbee, 2020a, 2020b; Park and Runkle, 2017; Zhen and van Iersel, 2016). However, the
inclusion of FR radiation in the range of PAR is yet to be widely accepted by the greater
scientific community. For decades, however, it has been broadly understood that FR radiation is
a predominant influencer of plant morphology and development (Elkins and van Iersel, 2020;
Owen et al., 2018; Park and Runkle, 2017; Craig and Runkle, 2013).
Photomorphogenic responses such as internode elongation, leaf expansion, and flowering
are regulated by various photoreceptors within plant cells including cryptochromes, phototropins,
and phytochromes (Hernández and Kubota, 2017; Park and Runkle, 2017; Poel and Runkle,
2017). For instance, a decreasing ratio of red (R; 600‒700 nm) and FR radiation emitted from a
radiation source generally promotes extension growth (Owen et al., 2018; Craig and Runkle,
2013), which is a function of phytochrome (Craig and Runkle, 2012; Borthwick and Hendricks,
1960). The influence of R and FR radiation on crop morphology is well documented. For
instance, Elkins and van Iersel (2020) reported that the height of foxglove ‘Dalmatian Peach’
(Digitalis purpurea) cut flower seedlings grown under sole-source lighting for 16 h·d–1 increased
by 38% as the R to FR ratio decreased from 13.7 to 0.6.
Phytochrome photoreceptors exist in two reversible conformations; PR and PFR. These
conformations are designated as the "inactive" and "active" conformations, respectively (Poel
and Runkle, 2017; Craig and Runkle, 2016), as PFR is primarily responsible for initiating
phytochrome-mediated photomorphogenic responses (Hernández and Kubota, 2017). The ratio
126
of R:FR radiation in an radiation source's spectrum can influence the ratios of these
conformations. When exposed to R radiation, PR changes conformation to PFR, while PFR can
revert to PR in the presence of either FR radiation or through natural degradation (Sager et al.,
1988). The ratio of these phytochrome conformations is referred to as the phytochrome
photoequilibrium (PPE; PFR/PR+FR), and it is closely associated with the activity of phytochromes
within plant cells (Craig and Runkle, 2016; Sager et al., 1988).
R and FR radiation are not only prominent drivers of crop architecture; they are also
integral to the flowering responses of many long-day plants (LDPs; Walters et al., 2019; Runkle
and Heins, 2001). When grown under a FR-radiation deficient filter, flowering of campanula
‘Blue Clips’ (Campanula carpatica), coreopsis ‘Early Sunrise’ (Coreopsis ×grandiflora) and
pansy ‘Crystal Bowl Yellow’ (Viola ×wittrockiana) was delayed by 2, 14, and 21 d,
respectively, compared to plants grown under a neutral filter that allowed for the transmission of
FR radiation (Runkle and Heins, 2001). It has also been shown that SL emitting moderate
intensities of FR radiation (≥15 µmol∙m–2∙s–1) can hasten flowering in LDPs compared to SL
without FR radiation (Kohler and Lopez, 2021). For instance, the LDP snapdragon ‘Liberty
Classic Yellow’ (Antirrhinum majus) grown under SL containing 15 µmol∙m–2∙s–1 of FR
radiation for a 16 h·d–1 for 28 d during the plug stage reached open flower 6 d faster than plants
grown under SL containing only blue (B) and R radiation during the plug stage (Kohler and
Lopez, 2021).
B radiation (400‒500 nm) inhibits extension growth in many crops, which is a function
of cryptochrome and phototropin photoreceptors (Park and Runkle, 2018; Fankhauser and
Christie, 2015; Wollaeger and Runkle, 2015). However, B radiation mediated stem compaction
responses are species-specific, and some crops defy this phenomenon (Hernández and Kubota,
127
2017). In a 2017 study, Poel and Runkle reported that geranium ‘Pinto Premium Salmon’
(Pelargonium ×hortorum) and petunia ‘Single Dreams White’ (Petunia ×hybrida) grown under
(%) B45R55 LEDs emitting a photosynthetic photon flux density (PPFD) of 90 ± 10 µmol∙m–2∙s–1
for 16 h·d–1 were ≈17% and 22% shorter, respectively, than those grown under SL provided by
B10G5R85 LEDs (Poel and Runkle, 2017). In a separate study, poinsettia ‘Christmas Spirit’ and
‘Christmas Eve’ (Euphorbia pulcherrima) grown under 100 ± 20 µmol·m–2·s–1 of high-pressure
sodium (HPS) SL with 5% B radiation for 10 h·d–1 were ≈52% and 36% taller, respectively, than
those grown under the same intensity and duration of SL provided by LEDs emitting 20% B
radiation for 12 weeks (Ashraful Islam et al., 2012).
Additionally, a moderate intensity of B radiation can function as a long-day signal for
some floriculture crops. For instance, a 4-h night interruption (NI) provided by 30 µmol·m–2·s–1
of B radiation was as effective as a 4-h NI provided by 2 µmol·m–2·s–1 from R+white (W)+FR
LEDs at promoting flowering in calibrachoa ‘Callie Yellow Improved’ (Calibrachoa ×hybrida),
coreopsis ‘Early Sunrise’, petunia ‘Wave Purple Improved’, rudbeckia ‘Indian Summer’
(Rudbeckia hirta), and snapdragon ‘Liberty Classic Yellow’ (Meng and Runkle, 2017).
Furthermore, SharathKumar et al. (2021) demonstrated the efficacy of a 4-h day extension (DE)
provided by 40 µmol·m–2·s–1 of 100% B radiation, creating a 15-h photoperiod, at inhibiting
flowering of greenhouse-grown chrysanthemum ‘Radost’ (Chrysanthemum morifolium).
Traditionally, high-intensity horticultural LED fixtures utilized a combination of B and R
diodes due to the higher absorption of B and R photons in upper leaf cells, consistent with the
peak absorbances of chlorophyll a and b, compared to other wavebands (Kusuma et al., 2021; Liu
and van Iersel, 2021). However, recent research has found that green radiation (G; 500‒600 nm)
is comparably effective for photosynthesis. For example, Liu and van Iersel (2021) reported that
128
whole-plant photosynthetic efficacy of G radiation applied to lettuce ‘Green Towers’ (Lactuca
sativa) was higher than that of B radiation when applied at intensities >500 µmol·m–2·s–1, as G
photons are transmitted further into the plant canopy than other wavebands (Kusuma et al., 2020;
Liu and van Iersel, 2021).
In addition to stimulating photosynthesis, G radiation has been shown to inhibit
branching of some ornamental plants when applied at moderate intensities (Meng and Runkle,
2019). For example, petunia ‘Easy Wave Burgundy Star’ had an average of ≈5 fewer lateral
branches when the G radiation photon flux density (PFD) during a 16-h DE was 25 µmol·m–2·s–1
compared to 2 µmol·m–2·s–1. Furthermore, moderate fluxes of G radiation can serve as a long-
day signal for some floriculture crops (Meng and Runkle, 2019). For example, G radiation
saturated the flowering response of ageratum ‘Hawaii Blue’ (Ageratum houstonianum) when
applied at intensities of 2 µmol·m–2·s–1 during a 16-h DE, although 13 µmol·m–2·s–1 was
required to saturate the flowering responses of petunia ‘Easy Wave Burgundy Star’ and ‘Wave
Purple Improved’, and snapdragon ‘Liberty Classic Yellow’ (Meng and Runkle, 2019).
Furthermore, in several studies on non-horticultural crops, G radiation inhibited some B
radiation-mediated photomorphogenic responses, such as hypocotyl compaction and anthocyanin
accumulation (Hernández and Kubota, 2017). In addition, certain fluxes of G radiation can elicit
stem elongation responses similar to that of FR radiation, which can be counteracted with B
radiation (Zhang et al., 2011). Interestingly, stem elongation of plants exposed to a combination
of G and FR radiation was greater than that of plants exposed to either waveband alone (Wang
and Folta, 2013).
When applied simultaneously, B, G, R, and FR wavebands can have compounding effects
on crop growth and development. For instance, height of high-wire cucumber ‘Elsie’ (Cucumis
129
sativus) and tomato ‘Climstar’ (Solanum lycopersicum) were up to ≈17% and 25% taller when
grown under 120 µmol·m–2·s–1 of B30G30R60 SL for 16 h·d–1 compared to the same intensity and
duration of B25R95 SL, suggesting that the addition of G radiation counteracted B-mediated plant
compaction, producing taller plants (Garcia and Lopez, 2020) Moreover, when B, G, and R
radiation is applied together, the resulting broad-spectrum radiation appears white (W) to the
human eye, increasing the visibility in the work environment. This can aid in detection of pests
and nutrient deficiencies compared to spectra comprised of one or two wavebands (Kusuma et
al., 2021).
Radiation quality can also influence flower petal color by influencing the accumulation of
pigments such as anthocyanins, carotenoids, and flavonoids (van Der Kooi et al., 2016; Zhao and
Tao, 2015). Petal color is influenced in part by petal morphology, i.e., tissue thickness and
inhomogeneity (van Der Kooi et al., 2016), which may be affected by radiation quality. While
flower color is of ecological importance to angiosperms as it helps attract specific pollinators
(van Der Kooi et al., 2016), it is also of significant aesthetic importance to consumers (Yue and
Behe, 2010). Manipulating radiation quality to produce cut flowers with more vibrant colors can
increase consumers’ willingness to buy and subsequent product enjoyment (Yue and Behe,
2010).
To our knowledge, minimal research examining the influence of supplemental radiation
quality on the greenhouse production of LDP specialty cut flowers has been published and thus,
additional research could provide utility to cut flower greenhouse growers. Therefore, the
objective of this study was to quantify the influence of SL radiation quality on time to flower and
harvest, and finished quality of three long-day specialty cut flowers. We hypothesized that
flowering would be delayed for plants grown under SL lacking FR radiation. Additionally, we
130
predicted that plants grown under R120 SL would exhibit greater stem elongation compared to the
other treatments, particularly B120 SL, where we predicted that stems would remain compact. We
also hypothesized that treatments with a combination of B, G, R, and FR wavebands would yield
shorter cut flowers when the emission spectrum contained a higher flux of B radiation, and
longer cut flowers when the emission spectrum contained a higher flux of FR radiation.
Materials and methods
Plant material, culture, and lighting treatments
Seeds of godetia ‘Grace Rose Pink’ (Clarkia amoena; Sakata Seed America, Morgan
Hill, CA), snapdragon ‘Potomac Royal’ (PanAmerican Seed, West Chicago, IL), and stock ‘Iron
Rose’ (Matthiola incana; Sakata Seed America) were sown in 162-cell trays at a commercial
propagator (Raker-Roberta’s Young Plants, Litchfield, MI). Three trays each of godetia,
snapdragon, and stock were received on 18 December 2020 [Replication (Rep.) 1] and 28
December 2021 (Rep. 2).
Young plants were grown in a glass-glazed greenhouse under a natural short-day
photoperiod with LED fixtures (Philips GP-TOPlight DRW-MB; Koninklijke Philips N.V.,
Eindhoven, Netherlands) providing a supplemental PPFD of ≈200 µmol·m–2·s–1 from 0730 to
1730 HR, creating a DLI of ≈15 mol·m–2·d–1. The greenhouse air average daily temperature
(ADT) set point was a constant 16 °C. Stock young plants were thinned after cotyledon
expansion to increase the amount of double flowering phenotypes, according to protocols
provided by the breeder (Sakata Ornamentals, 2022). Godetia and snapdragon young plants were
thinned upon cotyledon expansion. Young plants were irrigated as needed with MSU Plug
Special [13N–2.2P–10.8K water-soluble fertilizer containing (mg·L‒1) 61 nitrogen, 10
131
phosphorus, 50 potassium, 28.1 calcium, 4.7 magnesium, 1.3 iron, 0.6 manganese, 0.6 zinc, 0.6
copper, 0.4 boron, and 0.1 molybdenum; (GreenCare Fertilizers Inc., Kankakee, IL)] blended
with reverse-osmosis water and applied with a mist nozzle (Super Fine Fogg-It Nozzle; Fogg-It
Nozzle Co. Inc., Belmont, CA).
After 30 d under short days [18 January 2021 (Rep. 1) and 24 January 2022 (Rep. 2)],
180 godetia, snapdragon, and stock young plants were randomly selected for transplant. Seventy-
two bulb crates (39.3 cm wide × 59.7 cm long × 17.8 cm tall; 0.23 m2) were filled with a soilless
medium containing (by volume) 70% peat moss, 21% perlite, and 9% vermiculite (Suremix;
Michigan Grower Products Inc., Galesburg, MI). Each bulb crate held 10 young plants of an
individual genus, yielding 18 total bulb crates per genus. Young plants were transplanted at a
density of 43 plants per m2.
Three bulb crates of each genus were placed on benches on the ground in one of six
glass-glazed greenhouse compartments. High-intensity SL fixtures providing a total photon flux
density of 120 µmol·m–2·s–1 from 0700 to 1900 HR, creating a total DLI of ≈11 mol·m–2·d–1. This
was denoted as the vegetative stage. After four weeks, SL duration was increased to provide a
16-h photoperiod from 0600 to 2200 HR, creating a total DLI of ≈15 mol·m–2·d–1. This was
denoted as the reproductive stage. Whitewash (KoolRay Classic Liquid Shade, Continental
Products, Euclid, OH) and/or opaque black cloth covered compartment walls to prevent radiation
pollution between compartments and adjacent greenhouses. A quantum sensor (LI-190R, LI-
COR Biosciences, Lincoln, NE) positioned horizontally at plant height in each compartment
measured the PPFD every 10 s and a datalogger (CR1000; Campbell Scientific, Logan, UT)
recorded hourly averages. The actual DLIs during the vegetative and reproductive stages of the
two replications of the experiment were calculated and are provided in Tables 4.1 and 4.2.
132
SL treatments consisted of either 460-W HPS fixtures (LR48877; P.L. Light Systems,
Beamsville, ON, Canada), 631-W LED fixtures (VYPR 2p; Fluence, Austin, TX), 325-W LED
fixtures (LumiGrow Pro 325; LumiGrow, Emeryville, CA), 600-W LED fixtures (LX601G,
Heliospectra, Göteborg, Sweden), a combination of 72-W LED fixtures (HortiLED MULTI, P.L.
Light Systems) and 625-W LED fixtures (LumiGrow Pro 650E, LumiGrow), or 625-W LED
fixtures (LumiGrow Pro 650E; LumiGrow). SL treatments, defined by the PFD delivered at each
100-nm waveband of B (400‒500 nm), G (500‒600 nm), R (600‒700 nm), and FR (700‒800 nm)
radiation, were B7G60R44FR9 (HPS120), B20G50R45FR5, B20R85FR15, B30G25R65, B120, or R120,
respectively. The spectral distribution of the SL fixtures was measured at crop height in ten
random locations throughout each compartment with a spectrometer (LI-180; LI-COR
Biosciences) and are presented in Fig. 4.1. The PPE of each SL treatment was calculated
according to Sager et al. (1988) and are presented in Table 4.3.
Two layers of 15-cm supportive netting (HGN32804; Hydrofarm, Petaluma, CA) were
positioned ≈15 and ≈30 cm, respectively, above each bench. Greenhouse compartments were
equipped with evaporative-pad cooling and radiant hot water heating, which, in addition to
lighting fixtures, were controlled by an environmental control system (Priva Office version 725-
3030, Vineland Station, ON, Canada). The air ADT set points in each greenhouse compartment
were 15.8 °C (day/night 18.5/13 °C), with day temperatures maintained from 0800 to 1900 HR
and night temperatures maintained from 1900 to 0800 HR. An aspirated thermocouple [36-gauge
(0.127-mm diameter) type E, Omega Engineering, Stamford, CT] positioned in the middle of
each compartment measured the air temperature at plant height every 10 s, and the datalogger
recorded hourly means. Additionally, an infrared thermocouple (Type T, OS36-01; Omega
Engineering) positioned against an individual leaf of a snapdragon plant in each compartment
133
measured leaf temperature every 10 s, and the datalogger recorded hourly means. The actual air
ADTs, average daytime and nighttime temperatures at plant height, as well as average leaf
temperatures of each treatment during the vegetative and reproductive stages of the two reps. of
the experiment were calculated and are provided in Table 4.I.
Plants were irrigated as needed with MSU Orchid RO Special [13N–1.3P–12.5K water-
soluble fertilizer containing (mg∙L‒1) 125 nitrogen, 13 phosphorus, 121 potassium, 76 calcium,
19 magnesium, 1.7 iron, 0.4 copper and zinc, 0.9 manganese, 0.2 boron, and 0.2 molybdenum;
(GreenCare Fertilizers Inc.)] blended with reverse-osmosis water.
Data collection and analysis
Plants were monitored daily for the presence of first visible flower bud (VB) and first open
flower (OF). On the date of harvest (≥50 cm tall and three OFs for godetia; ≥50 cm tall and
inflorescence 50% open for snapdragon; ≥45 cm tall and inflorescence 50% open for stock),
stem length from the substrate surface to the tallest point of the inflorescence and caliper at the
thickest point of the stem [recorded with a digital caliper (3-inch carbon fiber digital caliper,
General Tools & Instruments, LLC, New York, NY)] were recorded for all plants. Additionally,
the total number of initiated inflorescences and branch number were recorded for snapdragon. A
colorimeter (CR-20 Color Reader; Konica Minolta Sensing, Inc., Chiyoda, Tokyo, Japan) was
utilized to measure flower petal color on three petals of each plant. Godetia flower color
measurements were taken on the pink portion of the flower petal interiors. Data were analyzed
using SAS (version 9.4; SAS Institute, Cary, NC) mixed model procedure (PROC MIXED) for
analysis of variance (ANOVA), and means were separated by Tukey’s honest significant
difference (HSD) test at P ≤0.05. SAS general linear models procedure (PROC GLM) was used to
134
fit regressions. Godetia and stock data were pooled across replications due to low harvestable
stem yield, and undetected single-flowering phenotypes being removed after transplant,
respectively.
Results
Time to visible flower bud
Time to VB (TVB) of godetia was influenced, albeit slightly, by SL spectrum. TVB was
the fastest for plants grown under HPS fixtures (52 d), and the slowest for plants grown under
R120 SL (56 d). TVB was similar under all other treatments (≈53 d; Fig. 4.2A). Snapdragon
grown under B20R85FR15, B120, and HPS120 SL reached VB the fastest (45–47 d), whereas TVB
was delayed by up to 10 and 4 d under R120 SL during reps. 1 and 2, respectively (Fig. 4.3A;
4.3B). TVB was delayed by 2‒4 d and 1‒3 d under B20G50R45FR5 and B30G25R65 SL,
respectively, compared to the fastest treatments. TVB of stock was the fastest for plants grown
under B120 SL (36 d). TVB was delayed by ≈2, 3, 3, and 3 d when grown under B20R85FR15,
B20G50R45FR5, B30G25R65, and HPS120 SL, respectively, compared to B120 SL. TVB was delayed
by 9 d for plants grown under R120 SL compared to B120 SL (Fig. 4.2B).
Time to open flower
Godetia time to OF (TOF) was the fastest for plants grown under HPS SL (79 d) and the
slowest for plants grown under B120 and R120 SL (88 and 91 d, respectively). TOF was similar for
all other SL treatments (84–86 d; Fig. 4.2C). Snapdragon reached OF the fastest when grown
under B20R85FR15 and B120 SL during rep. 1 (65 and 66 d, respectively; Fig. 4.3C; 4.3D).
However, HPS120 and B20R85FR15 SL hastened flowering the most during rep. 2 (68 and 70 d,
135
respectively). TOF was consistently delayed under R120 SL compared to the other treatments by
up to 19 and 8 d during reps. 1 and 2, respectively. Flowering was slightly delayed under
B20G50R45FR5 and B30G25R65 SL compared to the fastest treatments during both reps, although by
not as much as R120 SL. TOF was hastened for stock when grown under B120 and B20R85FR15 SL
(53 and 54 d, respectively). Flowering was delayed by 2‒3 d for plants grown under B30G25R65,
HPS120, and B20G50R45FR5 SL. Similar to TVB, TOF was delayed by 9 d when grown under R120
SL compared to B120 SL (Fig. 4.2D).
Time to harvest
Time to harvest (TTH) of godetia was the fastest under HPS120 SL and the slowest under
R120 SL (80 and 94 d, respectively). TTH was 85 d for plants grown under B20G50R45FR5,
B20R85FR15, and B30G25R65 SL, and 90 d for plants grown under B120 SL (Fig. 4.2E). TTH of
snapdragon was hastened when grown under B20R85FR15 and B120 SL during rep. 1 (67 and 69 d,
respectively; Fig. 4.3E), while TTH was fastest under HPS120 and B20R85FR15 SL during rep. 2
(69 and 72 d, respectively; Fig. 4.3F). TTH was slightly delayed when grown under
B20G50R45FR5 and B30G25R65 SL compared to the quickest treatments (4–5 d), while R120 SL
delayed harvest by up to 18 and 9 d during reps. 1 and 2, respectively. TTH of stock was the
fastest when grown under B120 and B20R85FR15 SL (54 and 55 d, respectively). Flowering was
delayed by ≈2 d for plants grown under B30G25R65, HPS120, and B20G50R45FR5 SL. TTH was
delayed by 10 d when grown under R120 SL compared to B120 SL (Fig. 4.2F).
136
Cut flower morphology at harvest
Godetia cut flower stems were the longest when grown under R120 SL and the shortest
when grown under B120, B20G50R45FR5, and HPS120 SL (124 and 109–113 cm, respectively; Fig.
4.2G). Godetia stem caliper was not influenced by SL treatment (data not reported). Snapdragon
stems were the shortest when grown under B120 SL. SL, regardless of rep (Fig. 4.3G; 4.3H).
Plants were ≈13 or 24 cm longer when grown under R120 SL during reps. 1 and 2, respectively.
During rep. 1, stems grown under B20R85FR15 SL were comparable in length to those grown
under B120 SL, although during rep. 2 they were ≈10 cm longer. Similarly, B30G25R65 SL
produced stems of similar length to B120 SL during rep. 1, while these stems were ≈17 cm longer
during rep. 2. Stems were of similar thickness regardless of SL treatment (data not reported).
Moreover, snapdragon grown under B20R85FR15, B120, and HPS120 SL had the fewest branches at
harvest (52–55 branches), while plants grown under R120 SL produced stems with 8‒11 more
branches (data not reported). The broad-spectrum LED fixtures produced stems with ≈5 fewer
branches than the R120 SL and up to 6 more branches than the other treatments (data not
reported). Snapdragon grown under R120, B120, and HPS120 SL had the fewest inflorescences at
harvest, while B20R85FR15 SL produced stems with ≈5 more inflorescences. B30G25R65 and
B20G50R45FR5 SL yielded stems with 1-2 fewer inflorescences than B20R85FR15 SL (data not
reported).
Stock stem length at harvest was commercially, but not statistically, similar between all
treatments. B120 and R120 SL produced the longest stems (53 to 54 cm), while B30G25R65 and
B20R85FR15 SL produced shorter cut flowers with an average stem length of ≈50 cm (Fig. 4.2H).
Stock stem caliper was similar for all treatments except B120, which produced stems up to 14%
thinner than the other treatments (data not reported).
137
Flower petal coloration at harvest
Godetia and stock flower petal coloration was not influenced by any SL treatment (Table
4.4). Snapdragon petal coloration was not commercially different between treatments.
Discussion
With a variety of commercially available SL fixtures on the market, it is important to
understand the influence that the supplemental radiation quality can have on the growth and
development of cut flowers. We found that development time, in addition to cut flower
morphology, varied between the spectra that were studied. Generally, TVB, TOF, and TTH were
the slowest for plants grown under R120 SL, regardless of variety. However, the varieties studied
exhibited different developmental responses to the remaining SL spectra. Godetia consistently
developed the fastest under HPS120 SL. Stock developed the fastest when grown under B120 and
B20R85FR15 SL, while snapdragon consistently developed the fastest when grown under
B20R85FR15, B120, and HPS120 SL.
While R radiation alone is sufficient to inhibit flowering in most short-day plants, many
LDPs require R and FR radiation, particularly when the DLI is low (e.g., <8 mol∙m–2∙d–1). Craig
and Runkle (2012) reported that flowering of snapdragon ‘Liberty Classic Cherry’ was delayed
by up to ≈14 d when grown under a 4-h NI provided by ≈1.5 µmol∙m–2∙s–1 of R radiation (PPE =
0.89) compared to the same NI provided by both R and FR radiation (PPE = 0.72). This
phenomenon may have contributed to the developmental delay seen under R120 SL across all
genera (Fig. 4.4; 4.5), which had an equivalent PPE of 0.89. While FR radiation from solar
radiation was available for plants under each treatment, SL emitting a moderate flux of FR
radiation reduced the estimated PPE and appeared to hasten plant development. The same is true
138
of the B120 SL treatment, which reduced the estimated PPE by 0.39 compared to the R120
treatment.
The effect of B120 SL on development time varied between the LDPs studied. While stock
and snapdragon experienced hastened development when grown under B120 SL, development of
godetia slowed when grown under B120 SL compared to most of the other treatments, indicating
that this response may be genus specific. This is supported by Hori et al., (2011), who reported
that baby’s breath ‘Bristol Fairy’ (Gypsophila paniculata) did not flower when grown under a
12-h DE provided by 20-30 µmol∙m–2∙s–1 of B radiation for 18 weeks. However, flowering
occurred after ≈75 or ≈98 d when plants were grown under 9 µmol∙m–2∙s–1 of incandescent
lighting or 20-30 µmol∙m–2∙s–1 of FR radiation for the same duration, respectively (Hori et al.,
2011). TVB, TOF, and TTH of stock and snapdragon was delayed as the estimated PPE
increased from 0.50 (B120) to 0.89 (R120; Fig. 4.4; 4.5), indicating that the developmental delay
between SL treatments could be at least partly due to increased phytochrome activity. This is in
agreement with Craig and Runkle (2016), who reported that TOF of the LDPs petunia 'Easy
Wave White' (Petunia ×hybrida) and snapdragon 'Liberty Classic Cherry' was delayed by up to
≈6 d and ≈12 d, respectively, as the estimated PPE of NI lighting increased from 0.46 to 0.89.
Both B20G50R45FR5 and B30G25R65 SL slightly delayed development compared to the
fastest treatments, however, not as significantly as R120 SL. This delay could be attributed to the
minimal emission of FR radiation in the former treatment and the lack of FR radiation in the
latter treatment, which resulted in higher estimated PPEs (0.85 to 0.87), although not as high as
R120 SL (0.89). Moreover, this delay may have lasted longer if these spectra did not contain B
and G radiation, as both wavebands can serve as long day signals when applied at moderate
intensities.
139
SL quality also influenced cut flower morphology. Stem lengths at harvest were generally
the shortest under B120 SL regardless of genus, and increased with the estimated PPE (Fig. 4.4;
4.5). Many floriculture crops exhibit a compact growth habit when grown under B radiation. For
instance, Zou (2018) found that geranium ‘Calliope Dark Red’ plants grown with 100% B
radiation for 24 h·d–1 was up to 6.7 cm wider in comparison to those grown with 100% R
radiation for 24 h·d–1. Moreover, baby’s breath ‘Bristol Fairy’ grown under a 24-h photoperiod
created with 16-h of DE lighting providing 20‒30 µmol∙m–2∙s–1 of B radiation was ≈43 cm
shorter than those grown under the same intensity and duration provided by 100% FR radiation
(Hori et al., 2011). This compaction could be at least partly regulated by phytochrome activity.
Kong et al. (2018) found that continuous exposure of 100 µmol∙m–2∙s–1 of B radiation for 14-20 d
promoted stem elongation of several bedding plants compared to the same intensity and duration
of R radiation. However, when ≈90 µmol∙m–2∙s–1 of B radiation was applied with an additional
flux of ≈10 µmol∙m–2∙s–1 of R radiation, plants were more compact than any other treatment. The
authors concluded that this response could be due to reduced phytochrome activity under sole-
source B radiation (PPE = 0.49), promoting stem elongation, compared to a combination of B
and R radiation (PPE = 0.74). Considering that plants in the present study were grown in
greenhouses with solar and supplemental radiation, the actual PPE under B120 SL would likely be
>0.50 due to the presence of other wavebands, potentially contributing to our similar findings.
However, stock cut flowers were the longest when grown under B120 and R120 SL and the
shortest when grown under B30G25R65 and B20R85FR15 SL, though differences were minimal.
This further supports the argument that B-mediated stem elongation is a genus-specific response.
Another instance of B-mediated stem elongation was published by Zou (2018), who found that
marigold ‘P-4’ (Tagetes erecta) grown under sole-source lighting providing 180 µmol∙m–2∙s–1 of
140
B radiation for 12 h·d–1 was up to 54% taller than those grown under 180 µmol∙m–2∙s–1 of R
radiation for the same duration. It was also found that petunia and dianthus seedlings grown
under SL emitting 19% B radiation for 16 h·d–1 were 59% and 3% taller than those grown under
SL emitting 6% B radiation (Collado and Hernández, 2022). However, the former SL treatment
included 5% G radiation, which may have antagonized B-mediated stem compaction compared
to the latter treatment, which did not contain G radiation.
Additionally, snapdragon grown under B20R85FR15 SL (estimated PPE = 0.84) had ≈5
more inflorescences at harvest compared to those grown under R120 SL (estimated PPE = 0.89).
This contrasts with Craig and Runkle (2012), who found that snapdragon ‘Liberty Classic
Cherry’ had eight more VBs when grown under 100% R NI lighting (estimated PPE = 0.89)
compared to other NI treatments creating an estimated PPE of 0.16 to 0.85.
The present study demonstrates the influence that SL quality can have on crop growth
and development. However, these effects cannot be relied on year-round as a means of crop
steering and growth regulation, as the effects of SL quality on crop growth and developmental
responses are the strongest when the solar DLI is low (Runkle et al., 2022; Runkle, 2017). For
instance, when the quotient of B radiation provided by SL increased from 0% to 30% when SL
provided 45-70% of the total DLI (ranging from 2.1‒8.4 mol·m–2·d–1), stem elongation of celosia
‘Fresh Look Gold’ (Celosia argentea), snapdragon ‘Rocket Pink’, and vinca ‘Titan Punch’
(Catharanthus roseus) was suppressed by ≈20%, ≈10%, and ≈30%, respectively (Runkle et al.,
2022; Randall and Lopez, 2014). In a separate study, where the DLI was consistently > 6.7
mol·m–2·d–1 and SL only provided 20-40% of the total DLI, there was no commercial effect on
seedling stem elongation as the quotient of B radiation provided by SL increased from 10% to
45% (Runkle et al., 2022; Poel and Runkle, 2017). Moreover, Hernández and Kubota (2012)
141
reported no statistical morphological differences between greenhouse-grown tomato seedlings
grown with SL of varying spectra and a DLI of either 8.9 or 19.4 mol·m–2·d–1. These findings
indicate that while SL spectrum may be less influential as the DLI increases, particularly >7
mol·m–2·d–1, it can have noticeable effects on crop growth and development when the solar DLI
is below this threshold.
While parameters including TOF, TTH, and finished stem quality must be considered
when selecting a spectrum for a SL strategy, human work suitability must also be considered. In
the present study, B20R85FR15, B120, and HPS120 SL consistently hastened plant development and
yielded cut flowers with moderate stem lengths. Conversely, R120 SL delayed TTH and produced
longer cut flowers. While a given spectrum may elicit desirable crop responses, it may create a
challenging work environment for humans by making it more difficult to diagnose cultural
issues, including nutrient deficiencies and pest prevalence on plant tissue (Kusuma et al., 2021).
This may be particularly true when the solar DLI is low, and SL contributes more to the total
DLI than solar radiation.
The color fidelity index (CFI; Rf) is an independent, unbiased indication of how well
natural colors can be perceived by the human eye under a particular light source (Kusuma et al.,
2021). The CFI exists on a scale of 0 to 100, where values closer to 100 indicate that the colors
perceived under a given light source are truer to nature (Kusuma et al., 2021). The CFI values of
each SL treatment were calculated with each source’s spectral power distribution according to
supplemental materials provided by IES (2020) and can be found in Table 4.3. While B20R85FR15
SL generally hastened TTH and produced stems with moderate lengths, it created an
environment with a lower CFI than HPS120 SL or either broad-spectrum fixture, meaning that
human visibility capacity would be impaired under that spectrum. However, the effects of
142
B20R85FR15 SL’s low CFI were the strongest during the early morning and evening, while solar
radiation was limiting. During the day, the higher fraction of solar radiation subjectively allowed
for sufficient human visibility. Additionally, both B120 (Rf <0) and R120 SL (Rf =33) created
environments that were inadequate for human visibility and sufficient crop supervision (Table
4.3). Similar to B20R85FR15 SL, the impact on visibility by these treatments was the strongest
when solar radiation was limiting; however, visibility was still noticeably impaired during the
day compared to any other treatments.
SL fixtures also vary in their capability to convert electrical power to photons. Photon
efficacy is defined as the number of moles of photons generated per energy input, typically
expressed as µmol∙J–1 (Katzin et al., 2021; Kusuma et al., 2020). Currently, LED fixtures can
have a photon efficacy of up to 3 µmol∙J–1, trumping the photon efficacy of HPS fixtures by
≈60%. This is partly because a substantial amount of energy consumed by HPS fixtures is
reemitted as heat, whereas LED fixtures typically function at a lower temperature. This can have
a significant impact on a greenhouse operation’s overall energy expenditures. While more energy
must be used to heat a greenhouse when using LED fixtures compared to HPS fixtures, the net
energy expenditure, and associated energy costs, can be 10‒25% lower than greenhouses
utilizing HPS fixtures (Katzin et al., 2021).
Rate of development, finished stem quality, crop visibility, and photon efficacy must be
considered when selecting a SL spectrum for one’s growing operation. Based on our findings, we
recommend utilizing an LED fixture that provides a light ratio similar to B20R85FR15 SL or
broad-spectrum light; both elicited desirable crop responses with minimal tradeoffs, while
allowing for sufficient human visibility. Although crops grown under HPS120 SL performed
143
similarly, we recommend utilizing LEDs as they most likely offer higher photon efficacy and the
potential for long-term energy and monetary savings.
Acknowledgements
We gratefully acknowledge BloomStudios, Sakata Seed America, and the Association of
Specialty Cut Flower Growers for funding, BloomStudios and Sakata Seed America for
providing seeds, and Fluence Bioengineering, Heliospectra, LumiGrow, and P.L. Light Systems
for LED fixtures. We thank Nate DuRussel, John Gove, and Ian Holcomb for greenhouse
assistance and data collection, Hydrofarm for netting, Raker Roberta’s Young Plants for sowing
seeds, and Syndicate Sales for floral supplies. This work was supported by the USDA National
Institute of Food and Agriculture, Hatch project MICL02472.
144
APPENDIX
145
Table 4.1. Actual daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], average daily temperatures (ADTs), mean day temperature,
mean night temperature, and mean leaf temperature [mean ± SD (°C)] for each supplemental light (SL) treatment during the vegetative
(VEG) and reproductive (REP) stages of replication 1. SL treatments consisted of either 460-W HPS fixtures (HPS120; LR48877; P.L.
Light Systems, Beamsville, ON, Canada), 631-W LED fixtures (B20G50R45FR5; VYPR 2p; Fluence, Austin, TX), 325-W LED fixtures
(B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville, CA), 600-W LED fixtures (B30G25R65; LX601G, Heliospectra, Göteborg,
Sweden), a combination of 72-W LED fixtures (B120; HortiLED MULTI, P.L. Light Systems) and 625-W LED fixtures (R120;
LumiGrow Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro 650E; LumiGrow).
SL Treatment DLI ADT Day Temperature Night Temperature Leaf Temperature
and Stage [mean ± SD (mol·m–2·d–1)] [mean ± SD (°C)] [mean ± SD (°C)] [mean ± SD (°C)] [mean ± SD (°C)]
HPS120
VEG 10.7 ± 2.0 15.6 ± 0.6 18.4 ± 1.6 12.7 ± 0.8 17.9 ± 3.5
REP 15.7 ± 4.5 16.1 ± 1.9 18.6 ± 2.7 13.5 ± 3.2 18.3 ± 3.4
B20G50R45FR5
VEG 10.8 ± 2.0 15.6 ± 0.5 18.3 ± 0.8 12.9 ± 1.0 17.0 ± 2.7
REP 15.6 ± 5.0 16.2 ± 1.7 18.9 ± 2.5 13.4 ± 3.1 18.4 ± 3.6
B20R85FR15
VEG 10.6 ± 2.0 15.6 ± 0.4 18.2 ± 0.8 12.9 ± 0.7 17.1 ± 2.9
REP 15.4 ± 8.9 16.2 ± 1.8 18.8 ± 2.5 13.6 ± 1.3 19.9 ± 4.4
B30G25R65
VEG 10.9 ± 2.1 15.5 ± 0.4 18.1 ± 0.8 12.9 ± 0.6 17.6 ± 2.9
REP 15.1 ± 4.2 16.2 ± 1.8 18.9 ± 2.6 13.5 ± 3.2 18.3 ± 2.7
B120
VEG 10.8 ± 2.1 16.1 ± 0.5 18.8 ± 1.3 13.3 ± 0.8 17.7 ± 2.4
REP 15.0 ± 4.1 16.5 ± 1.6 19.3 ± 2.5 13.7 ± 2.9 19.3 ± 3.7
R120
VEG 11.5 ± 2.5 15.5 ± 0.8 18.4 ± 2.6 12.6 ± 0.9 16.5 ± 3.8
REP 15.9 ± 5.3 16.0 ± 1.8 18.3 ± 3.1 13.8 ± 2.8 17.9 ± 3.5
146
Table 4.2. Actual daily light integrals (DLIs) [mean ± SD (mol·m–2·d–1)], average daily temperatures (ADTs), mean day temperature,
mean night temperature, and mean leaf temperature [mean ± SD (°C)] for each supplemental light (SL) treatment during the vegetative
(VEG) and reproductive (REP) stages of replication 2. SL treatments consisted of either 460-W high-pressure sodium fixtures (HPS120;
LR48877; P.L. Light Systems, Beamsville, ON, Canada), 631-W light-emitting diode (LED) fixtures (B20G50R45FR5; VYPR 2p;
Fluence, Austin, TX), 325-W LED fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville, CA), 600-W LED fixtures
(B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a combination of 72-W LED fixtures (B120; HortiLED MULTI, P.L. Light
Systems) and 625-W LED fixtures (R120; LumiGrow Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro 650E;
LumiGrow).
SL Treatment DLI ADT Day Temperature Night Temperature Leaf Temperature
and Stage [mean ± SD (mol·m–2·d–1)] [mean ± SD (°C)] [mean ± SD (°C)] [mean ± SD (°C)] [mean ± SD (°C)]
HPS120
VEG 11.7 ± 2.2 16.0 ± 0.5 18.7 ± 2.1 13.3 ± 0.8 18.0 ± 4.0
REP 16.1 ± 4.6 16.0 ± 1.0 18.9 ± 2.1 12.9 ± 1.5 18.7 ± 4.0
B20G50R45FR5
VEG 10.9 ± 2.8 16.3 ± 0.7 19.0 ± 2.4 13.6 ± 0.8 18.0 ± 3.1
REP 16.2 ± 5.2 15.9 ± 0.8 18.9 ± 2.2 12.9 ± 1.1 18.3 ± 3.2
B20R85FR15
VEG 11.6 ± 2.6 16.0 ± 0.5 18.7 ± 2.1 13.3 ± 0.8 18.3 ± 3.9
REP 15.9 ± 3.9 15.9 ± 1.0 18.9 ± 2.4 12.9 ± 1.3 18.5 ± 3.7
B30G25R65
VEG 11.6 ± 2.6 16.3 ± 0.7 19.0 ± 2.4 13.6 ± 0.8 18.0 ± 3.8
REP 16.4 ± 4.3 16.3 ± 1.4 19.3 ± 2.7 13.3 ± 1.7 18.0 ± 3.1
B120
VEG 11.6 ± 2.3 16.0 ± 1.0 19.1 ± 2.5 12.7 ± 1.2 17.9 ± 3.3
REP 15.9 ± 5.1 15.9 ± 0.9 18.9 ± 2.3 13.0 ± 1.6 18.3 ± 3.0
R120
VEG 11.8 ± 2.6 16.0 ± 1.0 19.1 ± 2.4 13.6 ± 0.8 17.8 ± 3.4
REP 15.9 ± 6.4 16.2 ± 1.4 19.5 ± 2.3 13.4 ± 1.2 18.3 ± 3.4
147
Table 4.3. Estimated phytochrome photoequilibria (PPE; PFR/PR+FR) and color fidelity index
(CFI; Rf) of each supplemental lighting (SL) treatment. PPEs were calculated according to Sager
et al. (1988) and CFI values were calculated according to supplemental materials provided by
IES (2018). SL treatments consisted of either 460-W HPS fixtures (HPS120; LR48877; P.L. Light
Systems, Beamsville, ON, Canada), 631-W LED fixtures (B20G50R45FR5; VYPR 2p; Fluence,
Austin, TX), 325-W LED fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville,
CA), 600-W LED fixtures (B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a
combination of 72-W LED fixtures (B120; HortiLED MULTI, P.L. Light Systems) and 625-W
LED fixtures (R120; LumiGrow Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro
650E; LumiGrow).
SL Treatment
HPS120 B20G50R45FR5 B20R85FR15 B30G25R65 B120 R120
Estimated PPE 0.85 0.85 0.84 0.87 0.50 0.89
CFI (Rf) 44 80 0 55 <0 33
148
Table 4.4. Adjusted hue angle (h°), chroma (C), and Hunter CIELAB (L*, a*, b*) values at
harvest for godetia, snapdragon, and stock grown under six different supplemental lighting (SL)
treatments. SL treatments consisted of either 460-W HPS fixtures (HPS120; LR48877; P.L. Light
Systems, Beamsville, ON, Canada), 631-W LED fixtures (B20G50R45FR5; VYPR 2p; Fluence,
Austin, TX), 325-W LED fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville,
CA), 600-W LED fixtures (B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a
combination of 72-W LED fixtures (B120; HortiLED MULTI, P.L. Light Systems) and 625-W
LED fixtures (R120; LumiGrow Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro
650E; LumiGrow). Letters indicate mean separations across treatments using Tukey-Kramer
honestly significant difference (HSD) test at P ≤0.05.
SL Treatment
Parameter HPS120 B20G50R45FR5 B20R85FR15 B30G25R65 B120 R120
Godetia ‘Grace Rose Pink’ NS
h° 355.2 354.9 354.5 354.3 353.6 355.2
C 48.5 49.1 47.7 49.4 47.4 50.1
L* 43.2 42.8 44.8 42.5 44.5 43.8
a* 48.3 48.9 47.5 49.1 47.1 49.9
b* -3.9 -4.2 -4.5 -4.4 -5.2 -3.7
Snapdragon ‘Potomac Royal’
h° 355.1
356.8 A 357.2 A 356.3 AB 356.9 A 355.7 BC
C
C 34.4 A 34.2 AB 33.0 BC 33.7 ABC 32.7 C 34.3 ABC
L* 19.4 B 19.7 B 19.6 B 20.2 B 19.8 B 21.7 A
a* 34.5 A 34.2 AB 33.0 BC 33.7 ABC 32.7 C 34.2 ABC
b* -2.7 D -1.8 BC -0.9 A -1.8 B -1.3 AB -2.6 CD
Stock ‘Iron Rose’ NS
h° 336.8 337.3 337.0 336.9 337.5 337.0
C 49.6 49.2 49.5 49.1 49.9 49.8
L* 33.9 33.9 34.1 33.6 34.1 33.3
a* 45.6 45.3 45.6 45.1 46.0 45.8
b* -19.5 -19.0 -19.3 -19.3 -19.1 -19.5
149
Table 4.5. Regression equations and R2 for time to visible bud, time to open flower, time to
harvest, and stem length at harvest of godetia 'Grace Rose Pink', stock 'Iron Rose', and
snapdragon 'Potomac Royal' in response to the estimated phytochrome photoequilibrium of each
supplemental lighting treatment. ** and *** indicate model significance at P <.001 and P
<.0001, respectively. All models are in the form of: ƒ = y0 + a*PPE + b*PPE2.
Parameter y0 a b R2
Godetia 'Grace Rose Pink'
Time to visible bud (d) 136.93z -265.74 196.25 0.206***
Time to open flower (d) 324.77 -746.35 543.98 0.371***
Time to harvest (d) 355.71 -841.28 613.57 0.400***
Stem length at harvest (cm) 352.60 -785.44 593.81 0.151***
Stock 'Iron Rose'
Time to visible bud (d) 163.75 -410.27 309.94 0.460***
Time to open flower (d) 203.47 -485.22 366.37 0.503***
Time to harvest (d) 212.48 -510.96 385.82 0.422***
Stem length at harvest (cm) 104.79 -160.00 113.78 0.081***
Snapdragon 'Potomac Royal'
Time to visible bud (d)
Rep. 1 220.56 -563.92 423.74 0.621***
Rep. 2 128.57 -259.47 193.97 0.334***
Time to open flower (d)
Rep. 1 429.16 -1166.49 874.18 0.782***
Rep. 2 221.23 -469.85 344.12 0.380***
Time to harvest (d)
Rep. 1 410.51 -1097.96 822.36 0.681***
Rep. 2 232.81 -500.12 366.46 0.398***
Stem length at harvest (cm)
Rep. 1 362.45 -760.81 586.91 0.268***
Rep. 2 264.58 -424.62 333.72 0.112**
z
Coefficients for model equations were used to generate Figs. 4.4 and 4.5.
150
5
Photon flux density (μmol·m–2·s–1·nm–1)
B20G50R45FR5
4 HPS120
B20R85FR15
B30G25R65
3 B120
R120
2
1
0
400 500 600 700 800
Wavelength (nm)
Figure 4.1. Emission spectra of supplemental lighting (SL) fixtures utilized throughout the
study. SL treatments consisted of either 460-W HPS fixtures (HPS120; LR48877; P.L. Light
Systems, Beamsville, ON, Canada), 631-W LED fixtures (B20G50R45FR5; VYPR 2p; Fluence,
Austin, TX), 325-W LED fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville,
CA), 600-W LED fixtures (B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a
combination of 72-W LED fixtures (B120; HortiLED MULTI, P.L. Light Systems) and 625-W
LED fixtures (R120; LumiGrow Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro
650E; LumiGrow).
151
Figure 4.2. Time to visible flower bud, time to open flower, time to harvest, and stem length at
harvest of godetia ‘Grace Rose Pink’ and stock ‘Iron Rose’ in response to SL spectrum, pooled
over two replications. SL treatments consisted of either 460-W HPS fixtures (HPS120;
LR48877; P.L. Light Systems, Beamsville, ON, Canada), 631-W LED fixtures
(B20G50R45FR5; VYPR 2p; Fluence, Austin, TX), 325-W LED fixtures (B20R85FR15;
LumiGrow Pro 325; LumiGrow, Emeryville, CA), 600-W LED fixtures (B30G25R65; LX601G,
Heliospectra, Göteborg, Sweden), a combination of 72-W LED fixtures (B120; HortiLED
MULTI, P.L. Light Systems) and 625-W LED fixtures (R120; LumiGrow Pro 650E,
LumiGrow), or 625-W LED fixtures (LumiGrow Pro 650E; LumiGrow).Letters indicate mean
separations across treatments using Tukey-Kramer honestly significant difference (HSD) test at
P ≤0.05. Bars represent the mean and error bars indicate standard error.
152
Figure 4.3. Time to visible flower bud, time to open flower, time to harvest, and stem length at
harvest of snapdragon ‘Potomac Royal’ in response to SL spectrum over two replications. SL
treatments consisted of either 460-W HPS fixtures (HPS120; LR48877; P.L. Light Systems,
Beamsville, ON, Canada), 631-W LED fixtures (B20G50R45FR5; VYPR 2p; Fluence, Austin, TX),
325-W LED fixtures (B20R85FR15; LumiGrow Pro 325; LumiGrow, Emeryville, CA), 600-W
LED fixtures (B30G25R65; LX601G, Heliospectra, Göteborg, Sweden), a combination of 72-W
LED fixtures (B120; HortiLED MULTI, P.L. Light Systems) and 625-W LED fixtures (R120;
LumiGrow Pro 650E, LumiGrow), or 625-W LED fixtures (LumiGrow Pro 650E; LumiGrow).
Letters indicate mean separations across treatments using Tukey-Kramer honestly significant
difference (HSD) test at P ≤0.05. Bars represent the mean and error bars indicate standard error.
153
Godetia 'Grace Rose Pink' Stock 'Iron Rose'
70
A B
60
Time to visible bud (d)
50
40
30
20
R2 = 0.206*** R2 = 0.460***
0
100
C D
Time to open flower (d)
80
60
40
R2 = 0.371*** R2 = 0.503***
0
100
E F
Time to harvest (d)
80
60
40
R2 = 0.400*** R2 = 0.422***
0
140 G H
Stem length at harvest (cm)
120
100
80
60
40
R2 = 0.151*** R2 = 0.081***
0
0.5 0.6 0.7 0.8 0.9 0.5 0.6 0.7 0.8 0.9
PFR/PR+PFR PFR/PR+PFR
Figure 4.4. Influence of estimated phytochrome photoequilibrium (PFR/PR+PFR) of supplemental
lighting treatments on time to visible bud, time to open flower, time to harvest, and stem length
at harvest of godetia 'Grace Rose Pink' and stock 'Iron Rose'. Black symbols represent means;
error bars represent standard error. R2 values are presented; ** and *** indicate model
significance at P <.001 and P <.0001, respectively. Coefficients are presented in Table 4.5.
154
Replication 1 Replication 2
70
A B
Time to visible bud (d)
60
50
40
30
R2 = 0.621*** R2 = 0.334***
0
C D
Time to open flower (d)
80
60
40
R2 = 0.782*** R2 = 0.380***
0
E F
Time to harvest (d)
80
60
40
R2 = 0.681*** R2 = 0.398***
0
G H
Stem length at harvest (cm)
160
140
120
100
R2 = 0.268*** R2 = 0.112**
0
0.5 0.6 0.7 0.8 0.9 0.5 0.6 0.7 0.8 0.9
PFR/PR+PFR PFR/PR+PFR
Figure 4.5. Figure 4.5. Influence of estimated phytochrome photoequilibrium (PFR/PR+PFR) of
supplemental lighting treatments on time to visible bud, time to open flower, time to harvest, and
stem length at harvest of snapdragon 'Potomac Royal' during replications 1 and 2. Black symbols
represent means; error bars represent standard error. R2 values are presented; ** and *** indicate
model significance at P <.001 and P <.0001, respectively. Coefficients are presented in Table
4.5.
155
REFERENCES
156
REFERENCES
Ashraful Islam, M., Kuwar, G., Clarke, J. L., Blystad, D-R., Gislerød, H. R., Olsen, J. E., Torre,
S., 2012. Artificial light from light-emitting diodes (LEDs) with a high portion of blue
light results in shorter poinsettias compared to high pressure sodium (HPS) lamps.
Scientia Hortic. 147, 136–143.
Collado, C., Hernández, R., 2022. Effects of Light Intensity, Spectral Composition,
and Paclobutrazol on the Morphology, Physiology, and Growth of Petunia, Geranium,
Pansy, and Dianthus Ornamental Transplants. J. Plant Growth Regulation, 41, 461–478.
Craig, D.S., Runkle, E.S., 2012. Using LEDs to quantify the effect of the red to far-red ratio of
night-interruption lighting on flowering of photoperiodic crops. Acta Hortic. 956, 179–
186.
Craig, D.S., Runkle, E.S., 2013. A moderate to high red to far-red light ratio from light-emitting
diodes controls flowering of short-day plants. J. Amer. Soc. Hortic. Sci. 138, 167–172.
Craig, D.S., Runkle, E.S., 2016. An intermediate phytochrome photoequilibria from night-
interruption lighting optimally promotes flowering of several long-day plants. Environ.
and Expt. Bot. 121, 132–138.
Elkins, C., van Iersel, M.W., 2020. Supplemental far-red light-emitting diode light increases
growth of foxglove seedlings under sole-source lighting. HortTechnology 30, 564‒569.
Fankhauser, C., Christie, J.M., 2015. Plant phototropic growth. Current Biol. 25, R384–R389.
Faust, J.E., Dole, J.M., 2021. Cut flowers and foliages. CAB International, Oxfordshire, United
Kingdom.
Faust, J.E., Logan, J., 2018. Daily light integral: a research review and high-resolution maps of
the United States. HortScience 53, 1250–1257.
Garcia, C., Lopez, R.G., 2020. Supplemental radiation quality influences cucumber, tomato, and
pepper transplant growth and development. HortScience 55, 804‒811.
Goins, G.D., Yorio, N.C., Sanwo, M.M., Brown, C.S., 1997. Photomorphogenesis,
photosynthesis, and seed yield of wheat plants grown under red light-emitting diodes
(LEDs) with and without supplemental blue lighting. J. Exp. Bot. 48, 1407–1413.
Hernández, R., Kubota, C., 2012. Tomato seedling growth and morphological responses to
supplemental LED lighting red:blue ratios under varied daily solar light integrals. Acta
Hortic. 956, 187–194.
157
Hernández, R., Kubota, C., 2017. Light quality and photomorphogenesis. In: Lopez, R.G.,
Runkle, E.S. (Eds.). Light management in controlled environments. Meister Media
Worldwide, Willoughby, OH, pp. 29.
Hori, Y., Nishidate, K., Nishiyama, M., Kanahama, K., Kanayama, Y., 2011. Flowering and
expression of flowering-related genes under long-day conditions with light-emitting
diodes. Planta 234, 321–330.
Illuminating Engineering Society (IES). 2020. IES method for evaluating light source color
rendition (IES TM-30-20). IES, New York, NY.
Katzin, D., Marcelis, L.F.M., van Mourik, S., 2021. Energy savings in greenhouses by transition
from high-pressure sodium to LED lighting. Applied Energy 281, 116019.
Kohler, A.E., Lopez, R.G., 2021. Duration of light-emitting diode (LED) supplemental lighting
providing far-red radiation during seedling production influences subsequent time to
flower of long-day annuals. Scientia Hortic. 30, 109956.
Kong, Y., Stasiak, M., Dixon, M.A., Zheng, Y., 2018. Blue light associated with low
phytochrome activity can promote elongation growth as shade-avoidance response: A
comparison with red light in four bedding plant species. . Environ. Expt. Bot. 155, 345–
359.
Kusuma, P., Fatzinger, B., Bugbee, B., Soer, W., Wheeler, R., 2021. LEDs for extraterrestrial
agriculture: tradeoffs between color perception and photon efficacy (NASA/TM-
20210016720). National Aeronautics and Space Administration, Kennedy Space Center,
FL.
Kusuma, P., Pattison, P.M., Bugbee, B., 2020. From physics to fixtures to food: current and
potential LED efficacy. Hortic. Res. 56.
Liu, J., van Iersel, M.W., 2021. Photosynthetic physiology of blue, green, and red light: light
intensity effects and underlying mechanisms. Front. Plant Sci. 12, 619987.
Lopez, R.G., Runkle, E.S., 2008. Photosynthetic daily light integral during propagation
influences rooting and growth of cuttings and subsequent development of new guinea
impatiens and petunia. HortScience 43, 2052–2059.
Meng, Q., Runkle, E.S., 2017. Moderate-intensity blue radiation can regulate flowering, but not
extension growth, of several photoperiodic ornamental crops. Environ. Expt. Bot. 134,
12–20.
Meng, Q., Runkle, E.S., 2019. Regulation of flowering by green light depends on its photon flux
density and involves cryptochromes. Phys. Plantarum 166, 762–771.
158
Owen, W.G., Meng, Q., Lopez, R.G., 2018. Promotion of flowering from far-red radiation
depends on the photosynthetic daily light integral. HortScience 53, 465–471.
Owen, W.G., Hilligoss, A., Lopez, R.G., 2016. Late-season high tunnel planting of specialty cut
flowers in the midwestern United States influences yield and stem quality.
HortTechnology 26, 338–343.
Ortiz, M.A., Hyrczyk, K., Lopez, R.G., 2012. Comparison of high tunnel and field production of
specialty cut flowers in the Midwest. HortScience 47, 1265–1269.
Park, Y., Runkle, R.S., 2018. Far-red radiation and photosynthetic photon flux density
independently regulate seedling growth but interactively regulate flowering. Environ.
Expt. Bot. 155, 206–216.
Produce Marketing Association and Food Marketing Institute (PMAFMI), 2016. Trends in mass-
market floral. Produce Marketing Association, Newark, DE.
Paradiso, R., Proietti, S., 2021. Light-quality manipulation to control plant growth and
photomorphogenesis in greenhouse horticulture: the state of the art and the opportunities
of modern LED systems. J. Plant Growth Reg.
Park, Y., Runkle, E.S., 2017. Far-red radiation promotes growth of seedlings by increasing leaf
expansion and whole-plant net assimilation. Environ. Expt. Bot. 136, 41–49.
Poel, B. R., Runkle, E.S., 2017. Spectral effects of supplemental greenhouse radiation on growth
and flowering of annual bedding plants and vegetable transplants. HortScience 52, 1221–
1228.
Prinzing, D., 2021. New findings on consumer attitudes toward domestic & locally-grown cut
flowers. Slow Flowers Journal. (accessed 27 Apr. 2021).
Randall, W.C., Lopez, R.G., 2014. Comparison of supplemental lighting from high-pressure
sodium lamps and light-emitting diodes during bedding plant seedling production.
HortScience 49, 589–595.
Runkle, E.S., Heins, R.D., 2001. Specific functions of red, far red, and blue light in flowering
and stem extension of long-day plants. J. Amer. Soc. Hortic. Sci. 126, 275–282.
Runkle, E.S., Park, Y., Meng, Q., 2022. High photosynthetic photon flux density can attenuate
effects of light quality. Acta Hortic. 1337, 333–340.
Runkle, E.S., 2017. Does light quantity trump light quality? Greenhouse Prod. News 27, 38.
Runkle, E.S., 2019. Including far red in an LED lighting spectrum. Greenhouse Prod. News 29,
58.
159
Sager, J.C., Smith, W.O., Edwards, J.L., Cyr, K.L., 1988. Photosynthetic efficiency and
phytochrome photoequilibria determination using spectral data. Amer. Soc. Agric. Eng.
31, 1882–1889.
Sakata Ornamentals, 2022. Stock Iron culture sheet. .
SharathKumar, M., Heuvelink, E., Marcelis, L.F.M., van Ieperen, W., 2021. Floral induction in
the short-day plant chrysanthemum under blue and red extended long-days. Front. Plant
Sci. 11, 610041.
Van der Kooi, C.J., Elzen, J.T.M., Staal, M., Stavenga, D.G., 2016. How to color a flower: on the
optical principles of flower coloration. Proc. Biol. Sci. 283, 20160429.
Walters, K.J., Hurt, A.A., Lopez, R.G., 2019. Flowering, stem extension growth, and cutting
yield of foliage annuals in response to photoperiod. HortScience 54, 661–666.
Wang, Y., Folta, K.M., 2013. Contributions of green light to plant growth and development.
Amer. J. Bot. 100, 70–78.
Wollaeger, H.M., Runkle, E.S., 2015. Growth and acclimation of impatiens, salvia, petunia, and
tomato seedlings to blue and red light. HortScience 50, 522–529.
Yue, C., Behe, B.K., 2010. Consumer color preferences for single-stem cut flowers on calendar
holidays and noncalendar occasions. HortScience 45, 78–82.
Zhang, T., Maruhnich, S.A., Folta, K.M., 2011. Green light induces shade avoidance symptoms.
Plant Phys. 157, 1528–1536.
Zhao, D., Tao, J., 2015. Recent advances on the development and regulation of flower color in
ornamental plants. Front. Plant Sci. 6, 261.
Zhen, S., Bugbee, B., 2020a. Far-red photons have equivalent efficiency to traditional
photosynthetic photons: implications for redefining photosynthetically active radiation.
Plant Cell Environ. 43, 1259–1272.
Zhen, S., Bugbee, B., 2020b. Substituting far-red for traditionally defined photosynthetic photons
results in equal canopy quantum yield for CO2 fixation and increased photon capture
during long-term studies: implications for re-defining PAR. Front. Plant Sci. 11, 581156.
Zhen, S., van Iersel, M.W., 2016. Far-red light is needed for efficient photochemistry and
photosynthesis. J. Plant Phys. 209, 115–122.
Zhen, S., van Iersel, M.W., Bugbee, B., 2021. Why far-red photons should be included in the
definition of photosynthetic photons and the measurement of horticultural fixture
efficacy. Front. Plant Sci. 22.
160
Zou, Y., 2018. Light-emitting diode lighting quality effects on morphology, growth, flowering,
and carotenoid content of potted plant, cut flower, and medicinal flower production
(Masters Thesis). Univ. of California, Davis, Davis, CA.
161