CONTROLLED ENVIRONMENT PRODUCTION IMPACTS HYDROPONICALLY GROWN CULINARY HERB PHYSIOLOGY, BIOCHEMISTRY, AND CONSUMER PREFERENCE By Kellie Jean Walters A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Horticulture – Doctor of Philosophy 2020 i ABSTRACT PREFERENCE By Kellie Jean Walters CONTROLLED ENVIRONMENT PRODUCTION IMPACTS HYDROPONICALLY GROWN CULINARY HERB PHYSIOLOGY, BIOCHEMISTRY, AND CONSUMER An increased demand for a consistent year-round supply of fresh, high-quality, and locally grown produce and the mitigation of food deserts has spurred interest in controlled environment agriculture (CEA) in greenhouses and indoor production facilities. Although technology to manipulate light quanitity [radiation intensity and daily light integral (DLI)], mean daily temperature (MDT), and carbon dioxide (CO2) concentration exists, its utility is limited when growth, development, and biochemical responses are largely unknown for many specialty crops such as culinary herbs. Therefore, the research objectives were to 1) characterize historical controlled environment (CE) production trends and assess the current state of the United States (U.S.) hydroponics industry; and 2) quantify the growth, development, volatile organic compound (VOC) concentration, and consumer sensory preferences of finish-stage culinary herbs in response to DLI and MDT and of seedling responses to DLI and CO2. Our survey revealed large variation in production practices and technologies utilized. Growers identified research on manipulating the growing environment to improve flavor and the creation of production recipes taking multiple parameters into account as the highest priorities. Therefore, surface regression models characterizing growth and developmental responses of dill, parsley, purple basil, sage, spearmint, sweet basil, and watercress to MDT and DLI were developed to estimate yield and morphology within the experimental range tested. These data will serve as a foundation, allowing growers to calculate and implement the most advantageous growing ii environment by taking growth, development, and energy costs into account. Altering MDT and DLI during CEA production can also affect VOC production and subsequently, consumer sensory attributes of sweet basil. Increasing MDT from 23 to 36 ºC increased the concentrations of the phenylpropanoids eugenol and methyl chavicol and the terpenoid 1,8 cineole, but not linalool. However, the increase in MDT and VOC concentrations only influenced consumer appearance, texture, and color preference, but not aroma, flavor, or overall preference. We also quantified the extent radiation intensity and CO2 concentration during indoor seedling production influenced sweet basil yield and VOC concentration. Increasing radiation intensity from 100 to 600 µmol·m‒2·s‒1 increased 1,8 cineole, linalool, and eugenol concentrations. However, increased VOC concentrations were not correlated with increased consumer preference. Aftertaste, bitterness/sweetness, color, flavor, overall liking, and texture preference were highest when basil was grown under a radiation intensity of 200 µmol·m‒2·s‒1. This lead us to determine that consumers prefer the characteristic basil flavor made up of 1,8 cineole, eugenol, and linalool, which was not prevalent enough in basil grown under 100 µmol·m‒2·s‒1 but too high when grown under 400 and 600 µmol·m‒2·s‒1 leading to a lower consumer preference. Basil seedling fresh mass was 284% greater when grown under 600 compared to 100 µmol·m‒2·s‒1. After being transplanted into a common greenhouse environment, both an 80% increase in fresh mass and increased eugenol concentration persisted. By concentrating resources during high density production phases, costs can be spread across many plants. Taking planting density and production duration into account, the increased lighting cost per plant during propagation can be as little as 5% of the same lighting during finishing. Together, these data can be used to improve yield, morphology, color, and flavor, potentially improving energy-use efficiency and the economic feasibility of CE culinary herb production. iii This thesis is dedicated to Alex Renny, your love and support have been integral to the completion of this milestone. iv ACKNOWLEDGEMENTS To my advisor, Roberto Lopez, thank you for accepting me into your lab, for giving me opportunities to expand my research and professional development horizons, and for helping me become the scientist I am today. Your guidance has been instrumental in my accomplishments at MSU and in securing a faculty position at the University of Tennessee. Words cannot express my gratitude. My dissertation would not be as strong or as polished without my committee’s guidance. Erik Runkle and Jennifer Boldt, you have taught me to be intentional with every word, and encouraged me to delve deeper and ask why, thank you. Randy Beaudry, thank you for introducing me to GCMS! Even though the turbo pump kept exploding, I’m thankful you allowed me to come into your lab, experiment, and learn. Thank you for being so accommodating! I have also had many unofficial mentors throughout this experience. Bridget Behe, thank you for your guidance not only on human-related research, but also on navigating my PhD program and securing a faculty position. I think adding the human dimension to this research truly made it unique and useful. Though Chris Currey was my Master of Science advisor at Iowa State University, his mentorship and guidance have continued through my time at Michigan State. Thank you Currey for your continued support and collaboration! This past year I served in an administrative role with the FAST (Future Academic Scholars in Teaching) Fellowship Program mentoring Fellows in their teaching-as-research projects and professional development. The FAST program has been instrumental in expanding the way I approach both research and teaching. Rique Campa, thank you for giving me this experience and for being a wonderful mentor not only regarding scholarly teaching, but also for vi navigating academia. Also, a huge thank you to the FAST Fellows and Steering Committee, especially Megan Shiroda, Joelyn de Lima, and Stefanie Bair. The floriculture group has been a great group to be a part of! Firstly, I’d like to thank Sean Tarr for being my right hand man. You are a wonderful person, thank you for all of your help throughout the years! Thanks to Nate DuRussel and all of the undergraduate students for your support, the Lopez lab, Allison Hurt, Annika Kohler, Charlie Garcia, Anthony Soster, and Caleb Spall, the Runkle lab, Cathy Whitman, Yujin Park, William Meng, Mengzi Zhang, and Nathan Kelly, and the Warner lab, Ryan Warner, Qiuxia Chen, Keivan Bahmani, and Prabhjot Kaur for all of your help and friendship! Thank you to all of the graduate students in the Department of Horticulture, but especially Phil Engelgau and Patrick Abeli. Phil, thank you for answering all of my questions and for helping me navigate the lab! Patrick, thank you for being a wonderful roommate and keeping me sane throughout this experience. I’d also like to thank Brian Poel for not only his friendship, but for introducing me to the MSU cycling club. Through the cycling club I met a wonderful friend, Suzanne Slack, who I hope to continue collaborations with in the future, and the most wonderful man, Alex Renny, who I hope to share all of life’s experiences with. Grandma Kay Walters, thank you for always answering when I call. You are the most positive person I’ve ever met. I hope to be like you as I progress through life, spreading positivity wherever I go. Finally, to my mother, Barbara Wall, thank you for teaching me to work hard and to never give up. The perseverance you’ve instilled in me has been paramount in completing this milestone. vii TABLE OF CONTENTS LIST OF TABLES ___________________________________________________________ xii LIST OF FIGURES __________________________________________________________ xiv CHAPTER 1 ________________________________________________________________ 1 HISTORICAL, CURRENT, AND FUTURE PERSPECTIVES FOR CONTROLLED ENVIRONMENT HYDROPONIC FOOD CROP PRODUCTION IN THE UNITED STATES ____________________________________________________________________ 1 Abstract. _________________________________________________________________ 3 Historical Perspectives of U.S. CE and Hydroponic Production ____________________ 4 Current U.S. Hydroponic Crop Production Practices: A Survey ____________________ 8 Approach. _______________________________________________________________ 8 Results and Discussion ______________________________________________________ 9 Producers and production area. ______________________________________________ 9 Crops and production systems. ______________________________________________ 11 Nutrient and water management. ____________________________________________ 13 Growing environment: Monitoring and control. _________________________________ 16 Crop quality. ____________________________________________________________ 19 Research priorities. _______________________________________________________ 20 Future Perspectives and Conclusions _________________________________________ 21 APPENDICES ______________________________________________________________ 24 23 APPENDIX A: TABLES AND FIGURES _______________________________________ 24 APPENDIX B: HYDROPONIC PRODUCTION SURVEY ________________________ 37 APPENDIX C: INTERNAL REVIEW BOARD APPROVAL ______________________ 43 LITERATURE CITED ______________________________________________________ 45 CHAPTER 2 _______________________________________________________________ 53 MODELING GROWTH AND DEVELOPMENT OF HYDROPONICALLY GROWN DILL, PARSLEY, AND WATERCRESS IN RESPONSE TO PHOTOSYNTHETIC DAILY LIGHT INTEGRAL AND MEAN DAILY TEMPERATURE _______________ 53 Abstract. ________________________________________________________________ 55 Introduction ______________________________________________________________ 56 Materials and Methods _____________________________________________________ 60 Plant production. _________________________________________________________ 60 Growth data collection and analysis. _________________________________________ 62 Results __________________________________________________________________ 63 Dill. ___________________________________________________________________ 63 Parsley. ________________________________________________________________ 64 Watercress. _____________________________________________________________ 64 Discussion ________________________________________________________________ 65 Photosynthetic DLI. _______________________________________________________ 65 Temperature. ____________________________________________________________ 66 viii DLI and MDT interactions. _________________________________________________ 68 Modeling. _______________________________________________________________ 70 Conclusions ______________________________________________________________ 72 APPENDIX ________________________________________________________________ 74 LITERATURE CITED ______________________________________________________ 85 CHAPTER 3 _______________________________________________________________ 90 THE INFLUENCE OF AVERAGE DAILY TEMPERATURE AND DAILY LIGHT INTEGRAL ON THE GROWTH, DEVELOPMENT, BIOMASS PARTITIONING, AND COLOR OF PURPLE BASIL, SAGE, SPEARMINT, AND SWEET BASIL __________ 90 Abstract. ________________________________________________________________ 92 Introduction ______________________________________________________________ 93 Materials and Methods _____________________________________________________ 97 Plant production. _________________________________________________________ 97 Growth data collection and analysis. _________________________________________ 98 Results _________________________________________________________________ 100 Purple basil. ____________________________________________________________ 100 Sage. _________________________________________________________________ 102 Spearmint. _____________________________________________________________ 103 Sweet basil. ____________________________________________________________ 104 Discussion _______________________________________________________________ 104 Plant development. ______________________________________________________ 104 Growth, mass concentration, and partitioning. _________________________________ 106 Fresh mass modeling. ____________________________________________________ 108 Color. _________________________________________________________________ 110 Conclusions _____________________________________________________________ 111 APPENDIX _______________________________________________________________ 113 LITERATURE CITED _____________________________________________________ 131 CHAPTER 4 ______________________________________________________________ 136 LEVERAGING CONTROLLED-ENVIRONMENT AGRICULTURE TO INCREASE KEY BASIL TERPENOID AND PHENYLPROPANOID CONCENTRATIONS: THE EFFECTS OF RADIATION INTENSITY ON CONSUMER PREFERENCE ________ 136 Abstract. _______________________________________________________________ 138 Introduction _____________________________________________________________ 139 Materials and Methods ____________________________________________________ 143 Seedling production. _____________________________________________________ 143 VOC data collection and analysis. __________________________________________ 144 Sensory analysis. ________________________________________________________ 145 Statistical design and analysis. _____________________________________________ 146 Results _________________________________________________________________ 147 Seedling volatile organic compound concentrations. ____________________________ 147 Sensory panel. __________________________________________________________ 147 Comparing concentrations to sensory panel. __________________________________ 149 Discussion _______________________________________________________________ 150 Terpenoids. ____________________________________________________________ 150 ix Phenylpropanoids. _______________________________________________________ 151 Compound sensitivity. ____________________________________________________ 152 Consumer preferences. ___________________________________________________ 154 Comparing compound concentrations and preferences. __________________________ 155 Conclusions _____________________________________________________________ 156 APPENDICES _____________________________________________________________ 157 APPENDIX A: TABLES AND FIGURES ______________________________________ 157 158 APPENDIX B: CONSUMER SENSORY EVALUATION SURVEY ________________ 172 APPENDIX C: CONSUMER SENSORY EVALUATION INTERNAL REVIEW BOARD APPROVAL ______________________________________________________________ 180 LITERATURE CITED _____________________________________________________ 184 CHAPTER 5 ______________________________________________________________ 191 HYDROPONIC BASIL PRODUCTION: TEMPERATURE INFLUENCES THE PROFILE OF VOLATILE ORGANIC COMPOUNDS, BUT NOT OVERALL CONSUMER PREFERENCE ________________________________________________ 191 Abstract. _______________________________________________________________ 193 Introduction _____________________________________________________________ 194 Methods ________________________________________________________________ 197 Growing environment. ____________________________________________________ 197 GCMS analysis. _________________________________________________________ 197 Sensory analysis. ________________________________________________________ 197 Experimental design and statistical analysis. __________________________________ 198 Results _________________________________________________________________ 199 VOC concentration. ______________________________________________________ 199 Consumer preference. ____________________________________________________ 199 Comparison – Biplot and correlations. _______________________________________ 199 Discussion _______________________________________________________________ 200 Volatile organic compound concentration. ____________________________________ 200 Consumer preferences and correlations. ______________________________________ 202 Conclusions _____________________________________________________________ 204 APPENDIX _______________________________________________________________ 206 LITERATURE CITED _____________________________________________________ 214 CHAPTER 6 ______________________________________________________________ 218 SOLE-SOURCE RADIATION INTENSITY AND CARBON DIOXIDE CONCENTRATION DURING BASIL SEEDLING PRODUCTION INFLUENCE SUBSEQUENT YIELD AND FLAVOR COMPOUND CONCENTRATION DURING GREENHOUSE PRODUCTION _____________________________________________ 218 Abstract. _______________________________________________________________ 220 Introduction _____________________________________________________________ 221 Materials and Methods ____________________________________________________ 225 Seedling production. _____________________________________________________ 225 Finished plant production. ________________________________________________ 226 Growth, development, and VOC data collection and analysis. _____________________ 227 Statistical design and analysis. _____________________________________________ 228 x Results _________________________________________________________________ 228 Seedlings. ______________________________________________________________ 228 Harvest. _______________________________________________________________ 229 Finished volatile organic compound concentrations. ____________________________ 230 Discussion _______________________________________________________________ 230 Increased radiation intensity increased growth and morphological attributes. ________ 230 CO2 concentration did not influence mass. ____________________________________ 231 CO2 concentration influenced morphology. ___________________________________ 233 Seedling production conditions influence basil yield and quality at harvest. __________ 234 Efficiency implications. ___________________________________________________ 235 Conclusions _____________________________________________________________ 236 APPENDIX _______________________________________________________________ 237 LITERATURE CITED _____________________________________________________ 247 xi LIST OF TABLES Table I-1. The number of operations, area under protection, kilograms (kg) produced, total sales, and sales adjusted for inflation of food crops grown under protection in the United States (U.S.) between 1929 and 2014 as reported by the U.S. Department of Commerce or the U.S. Department of Agriculture. Values do not include mushroom production. ................................. 25 Table I-2. The types of nutrient solution filters used by respondents of a 2017 United States hydroponic grower survey. ........................................................................................................... 27 Table I-3. The target carbon dioxide (CO2) concentration as reported by respondents to a 2017 survey of hydroponic growers in the United States. ..................................................................... 28 Table II-1. Average daily light integral (DLI; mol·m‒2·d‒1 ± SD) and mean daily air (MDT), leaf, and nutrient solution temperature over the two (watercress), three (dill), or four (parsley) week growing period for two replications over time. Data were collected every 15 s with means logged every hour. .................................................................................................................................... 75 Table III-1. Average daily light integral (DLI; mol·m‒2·d‒1 ± SD) and mean daily air temperature (MDT), leaf, and nutrient solution temperatures (°C ± SD) over the three (spearmint and sweet basil), four (purple basil), or five (sage) week growing period for two replications over time. Plants were transplanted on 6 Sept. 2017 (rep 1, sweet basil), 20 Oct. 2017 (rep 2, sweet basil), 19 Apr. 2018 (rep 1, purple basil, sage, and spearmint), and 30 Oct. 2018 (rep 2, purple basil, sage, and spearmint). Data were collected every 15 s with means logged every hour. .............. 114 Table III-2. Regression analysis parameters and R2 or r2 for purple basil, sage, spearmint, and sweet basil branch number, height, maximum quantum yield of dark-adapted leaves (Fv/Fm), leaf area of four most recently fully expanded leaves, dry matter concentration (DMC), leaf mass fraction, and fresh mass in response to mean daily temperature (MDT; °C) and daily light integral (DLI; mol·m‒2·d‒1). All models are in the form of: ! = y0 + a*MDT + b*DLI + c*MDT2 + d*DLI2 + e*MDT*DLI. ........................................................................................................... 117 Table IV-1. Target radiation intensity, actual radiation intensity, and average daily air, canopy, and substrate temperatures (mean ± SD) during the seedling growth stage (2 weeks). .............. 159 Table V-1. Mean daily light integral (DLI; mol·m‒2·d‒1 ± SD) and mean daily air temperature (MDT), leaf, and nutrient solution temperatures during the three-week growing period for sweet basil (Ocimum basilicum ‘Nufar’) with two replications in over time. Data were collected every 15 s with means logged every hour. ............................................................................................ 207 Table VI-1. The date of sweet basil ‘Nufar’ (Ocimum basilicum) seed sowing, target and actual CO2 concentration (± SD), target and actual radiation intensity (± SD), and average daily air, canopy, and substrate temperature (± SD) during the seedling growth stage (2 weeks). ............ 238 xii Table VI-2. Actual average daily air and canopy temperature and daily light integral (DLI) (mean ± SD) during post-transplant greenhouse production (3 weeks of sweet basil ‘Nufar’ (Ocimum basilicum). ................................................................................................................... 239 xiii LIST OF FIGURES Figure I-1. The area of production dedicated to hydroponics per producer as reported by respondents to a 2017 hydroponic grower survey in the United States. n = 42 ............................ 29 Figure I-2. The proportion of crops grown based on number of producers growing the crop weighted by the percent value of that crop compared to total (left), and the percentage of producers growing each crop (right), based on respondents of a 2017 United States hydroponic grower survey. n = 42 ................................................................................................................... 30 Figure I-3. The proportion of hydroponic production systems used by respondents of a 2017 United States hydroponic grower survey, based on space dedicated to each system (left), and the percentage of respondents utilizing each type of production system based on frequency (right). n = 42 ............................................................................................................................................... 31 Figure I-4. The environmental and cultural parameters monitored by hydroponic growers in the United States during propagation (n = 38) and finished production (n = 39) based on responses to a 2017 survey of hydroponic growers in the United States. ......................................................... 32 Figure I-5. The frequency of nutrient solution replacement as reported by respondents of a 2017 United States hydroponic grower survey. n = 40 .......................................................................... 33 Figure I-6. The degree of benefit of research topics as reported by respondents to a 2017 survey of United States hydroponic growers. Means ± sd were calculated by assigning values to the responses, 1 = not at all beneficial, 2 = slightly beneficial, 3 = moderately beneficial, 4 = very beneficial. Research topics that share letters do not differ by Tukey’s honestly significant difference test at P ≤ 0.05. n = 36 ................................................................................................. 34 Figure I-7. The use of supplemental or sole-source lighting based on responses to a 2017 survey of hydroponic growers in the United States. n = 40 ..................................................................... 36 Figure II-1. Mean daily temperature (MDT) and daily light integral (DLI) effects on dill (Anethum graveolens) fresh mass (A) and height (C), watercress (Nasturtium officinale) maximum quantum yield of dark-adapted leaves (Fv/Fm; B), height (D), and branch number (E). Response surfaces represent model predictions. Coefficients for these models are presented in Table II 2. Models are each based on 270 individual measurements. .......................................... 77 Figure II-2. Mean daily temperature (MDT) effects on parsley (Petroselinum crispum) fresh mass (A), height (C), leaf number (E), and maximum quantum yield of dark-adapted leaves (Fv/Fm; G), watercress (Nasturtium officinale) fresh mass (B) and branch length (D), and dill (Anethum graveolens) leaf number (F). Lines represent model predictions, with the coefficients for these models presented in Table II-2. Symbols (means ± SD) represent measured data (n = 27). ................................................................................................................................................ 81 xiv Figure II-3. Mean daily temperature (MDT; A) and daily light integral (DLI; B) effects on dill (Anethum graveolens; ●), parsley (Petroselinum crispum; ), and watercress (Nasturtium officinale; coefficients for these models presented in Table II-2. Symbols (means ± SD) represent measured data used to generate the models (A, n = 27; B, n = 9). ................................................................ 83 ) dry matter concentration (DMC). Lines represent model predictions, with the Figure III-1. Mean daily temperature (MDT) effects on purple basil ‘Dark Opal’ (Ocimum basilicum) branch number (A), dry matter concentration (DMC; C), and fresh mass (E), and daily light integral (DLI) effects on branch number (B), DMC (D), and fresh mass (F). Lines represent model predictions, with the coefficients for these models presented in Table III-2. Symbols (means ± SD) represent measured data (A, C, E, n = 27; B, D, F, n = 9). .................... 119 Figure III-2. Mean daily temperature (MDT) and daily light integral (DLI) effects on purple basil ‘Dark Opal’ (Ocimum basilicum) height (A), maximum quantum yield of dark-adapted leaves (Fv/Fm; B), leaf area (of the four leaves measured; C), and leaf fresh mass fraction (D). Response surfaces represent model predictions. Coefficients for these models are presented in Table III-2. Models are each based on 300 individual measurements. .......................................................... 120 Figure III-3. Mean daily temperature (MDT) and daily light integral (DLI) effects on purple basil ‘Dark Opal’ (Ocimum basilicum) hue angle (h°; A) and L* (B). Response surfaces represent model predictions. Coefficients for these models are presented in Table III-2. Models are each based on 300 individual measurements. ..................................................................................... 121 Figure III-4. Mean daily temperature (MDT) and daily light integral (DLI) effects on purple basil ‘Dark Opal’ (Ocimum basilicum). Replication 1 (A) was harvested on 17 May 2018, and replication 2 (B) was harvested on 27 Nov. 2018, 7 weeks after sowing and 4 weeks after transplant. Images depict representatives of plants measured. ................................................... 122 Figure III-5. Mean daily temperature (MDT) and daily light integral (DLI) effects on sage ‘Extrakta’ (Salvia officinalis) branch number (A), height (B), leaf fresh mass fraction (C), and leaf fresh mass (D). Response surfaces represent model predictions. Coefficients for these models are presented in Table III-2. Models are each based on 300 individual measurements. 124 Figure III-6. Mean daily temperature (MDT) effects on sage ‘Extrakta’ (Salvia officinalis) dry matter concentration (DMC; A), leaf area (C), and maximum quantum yield of dark-adapted leaves (Fv/Fm; D), and daily light integral (DLI) effects on DMC (B). Lines represent model predictions, with the coefficients for these models presented in Table III-2. Symbols (means ± SD) represent measured data (A, C, D, n = 27; B, n = 9). ........................................................... 125 Figure III-7. Mean daily temperature (MDT) and daily light integral (DLI) effects on spearmint ‘Spanish’ (Mentha spicata) branch number (A), height (B), maximum quantum yield of dark- adapted leaves (Fv/Fm; C), leaf area (of the four leaves measured; D), dry matter concentration (DMC; E), leaf fresh mass fraction (F), and fresh mass (G). Response surfaces represent model xv predictions. Coefficients for these models are presented in Table III-2. Models are each based on 300 individual measurements. .................................................................................................... 126 Figure III-8. Mean daily temperature (MDT) and daily light integral (DLI) effects on sweet basil ‘Nufar’ (Ocimum basilicum) branch number (A), node number (B), height (C), leaf area (of the four leaves measured; D), and fresh mass (E). Response surfaces represent model predictions. Coefficients for these models are presented in Table III-2. Models are each based on 300 individual measurements. ........................................................................................................... 128 Figure III-9. Predicted optimal mean daily temperature (MDTopt) to achieve the greatest sage ‘Extrakta’ (Salvia officinalis; A), sweet basil ‘Nufar’ (Ocimum basilicum; C), and spearmint ‘Spanish’ (Mentha spicata; E) fresh mass based on actual daily light integral (DLI). Also, predicted DLIopt to achieve the greatest sage (B) and sweet basil (D) fresh mass based on actual MDT. Equations were generated based on a surface regression model (Figures 4D, 6G, and 7E) with model coefficients reported in Table III-2. ......................................................................... 130 Figure IV-1. Spectral quality of light-emitting diode (LED) fixtures providing 20:40:40 blue:green:red radiation ratios (%), a red:far-red ratio of 13:1, and target radiation intensities of 100, 200, 400, or 600 µmol·m‒2·s‒1. ........................................................................................... 160 Figure IV-2. Concentrations [ng·mg‒1 dry mass (DM)] of 1,8 cineole (A), linalool (B), eugenol (C), and methyl chavicol (D) of sweet basil ‘Nufar’ (Ocimum basilicum) seedlings grown under 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD) for two weeks. Each symbol represents the mean of 20 plants ± SE. Lines represent linear or quadratic regression. *** indicates significant at P ≤ 0.001. ..................................................................... 161 Figure IV-3. Mean overall liking of sweet basil ‘Nufar’ (Ocimum basilicum) grown under 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD), based on a 9-point Likert scale ranging from dislike extremely (1) to like extremely (9). Means not followed by the same letter are significantly different by Tukey’s honestly significant difference test (P< 0.05). Each symbol represents 90 responses ± SD. ................................................................................ 163 Figure IV-4. Mean flavor (A), aftertaste (B), and aroma (C) of sweet basil ‘Nufar’ (Ocimum basilicum) grown under 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD), based on a 9-point Likert scale ranging from dislike extremely (1) to like extremely (9) and the bitterness/sweetness (D) based on a 9-point Likert scale ranging from extremely bitter (1) to extremely sweet (9). Means not followed by the same letter are significantly different by Tukey’s honestly significant difference test (P< 0.05). Each symbol represents 90 responses ± SD. ............................................................................................................................................... 164 Figure IV-5. Leaves of basil used in the consumer sensory analysis panel. Sweet basil ‘Nufar’ (Ocimum basilicum) was grown under radiation intensities of 100, 200, 400, or 600 µmol·m–2·s–1 for a 16-h photoperiod to create daily light integrals of 6, 12, 23, or 35 mol·m‒2·d‒1 for two weeks after sowing. ..................................................................................................................... 166 xvi Figure IV-6. The mean appearance (A), color (B), leaf size (C), and texture (D) of sweet basil ‘Nufar’ (Ocimum basilicum) grown under 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD), based on a 9-point Likert scale ranging from dislike extremely (1) to like extremely (9). Means not followed by the same letter are significantly different by Tukey’s honestly significant difference test (P< 0.05). Each symbol represents 90 responses ± SD. ............................................................................................................................................... 167 Figure IV-7. Frequency heat map of words used to describe sweet basil ‘Nufar’ (Ocimum basilicum) grown under radiation intensities of 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD). Chi2 describes the goodness of fit, phylogenetic trees depict the cluster analysis relationship between words used and the relationship between words used and samples. All words reported exhibited differences P< 0.05 based on 90 respondents. ................................................................................................................................ 169 Figure IV-8. Principal component analysis (PCA) showing the biplot differentiation of sweet basil ‘Nufar’ (Ocimum basilicum) grown under 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD), based on consumer sensory preferences (n = 90) and the concentration of two terpenoids and two phenylpropanoids (n = 20). ................................. 170 Figure IV-9. Heat map illustrating the correlation between volatile organic compounds (VOCs) and sensory preference characteristics of sweet basil ‘Nufar’ (Ocimum basilicum) grown under radiation intensities of 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD). Blue and red represent negative and positive correlations, respectively. Asterisks indicate significant correlations based on Pearson’s correlation *, P< 0.05; **, P<0.01. .......... 171 Figure V-1. Concentrations [ng·mg‒1 dry mass (DM)] of 1,8 cineole (A), linalool (B), eugenol (C), and methyl chavicol (D) of sweet basil (Ocimum basilicum ‘Nufar’) grown at mean daily temperatures (MDT) from 23 to 36 °C for three weeks. Symbols represent the mean of 15 plants ± SE. Lines represent linear regression. * and *** indicate significant at P ≤ 0.05 or 0.001, respectively. ................................................................................................................................ 208 Figure V-2. Appearance (A), texture (B), color (C), and bitterness/sweetness (D) of sweet basil (Ocimum basilicum ‘Nufar’) grown at mean daily temperatures (MDT) from 25 to 35 °C for three weeks, based on a 9-point Likert scale ranging from dislike extremely (1) to like extremely (9). Means not followed by the same letter are significantly different by Tukey’s honestly significant difference test (P < 0.05). Each bar represents 86 responses ± SD. .......................... 211 Figure V-3. Principal component analysis (PCA) showing biplot differentiation of sweet basil (Ocimum basilicum ‘Nufar’) grown at mean daily temperatures (MDT) from 23 to 36 °C for three weeks, based on consumer sensory preferences (n = 86) and the concentration of two terpenoids and two phenylpropanoids (n = 15). ......................................................................... 212 Figure V-4. Heat map illustrating the correlation between volatile organic compounds (VOCs) and sensory preference characteristics of sweet basil (Ocimum basilicum ‘Nufar’) grown at mean xvii daily temperatures (MDT) from 23 to 36 °C for three weeks. Blue and red represent negative and positive correlations, respectively. Asterisks indicate significant correlations based on Pearson’s correlation *, P < 0.05; **, P < 0.01. .......................................................................................... 213 Figure VI-1. Spectral quality of broad-spectrum light-emitting diode (LED) fixtures providing 20:40:40 blue:green:red radiation ratios (%), a red:far-red ratio of 13:1, and target radiation intensities of 100, 200, 400, or 600 µmol·m–2·s–1. ..................................................................... 240 , 1,000 µmol·mol–1; Figure VI-2. Radiation intensity (100, 200, 400, or 600 µmol·m–2·s–1) for a 16-h photoperiod to create daily light integrals of (6, 12, 23, or 35 mol·m‒2·d‒1) and CO2 concentration ( , 500 , pooled) effects on sweet basil ‘Nufar’ (Ocimum basilicum) µmol·mol–1; seedling height (A), dry mass (B), leaf area (C), fresh mass (D), and stem width (E) two weeks after sowing. Lines represent linear or quadratic regressions. Symbols (means ± SE) represent measured data ( and or 0.001, respectively. ................................................................................................................. 241 , n = 30; , n = 60). *, **, and *** indicate significant at P ≤ 0.05, 0.01, , 1,000 µmol·mol–1; Figure VI-3. Radiation intensity (100, 200, 400, or 600 µmol·m–2·s–1) for a 16-h photoperiod to create daily light integrals of (6, 12, 23, or 35 mol·m‒2·d‒1) and CO2 concentration ( , 500 µmol·mol–1; , pooled) administered during seedling production, two weeks after sowing. The figures depict seedling treatment effects on sweet basil ‘Nufar’ (Ocimum basilicum) height (A), branch number (B), stem width (C), node number (D), dry mass (E), and fresh mass (F) three weeks after transplant into a common enviornment. Lines represent linear or quadratic regressions. Symbols (means ±SE) represent measured data ( and , n = 30; , n = 60). * and *** indicate significant at P ≤ 0.05 or 0.001, respectively. .............................. 244 Figure VI-4. Concentrations [ng·mg‒1 dry mass (DM)] of 1,8 cineole (A), linalool (B), eugenol (C), and methyl chavicol (D) of sweet basil ‘Nufar’ (Ocimum basilicum) seedlings grown under radiation intensity (100, 200, 400, or 600 µmol·m–2·s–1) for a 16-h photoperiod to create daily light integrals of (6, 12, 23, or 35 mol·m‒2·d‒1) for two weeks and then transplanted in a common greenhouse environment and grown for three weeks. Each symbol represents the mean of 20 plants ± SE. Lines represent linear or quadratic regression. ** indicates significant at P ≤ 0.01. ..................................................................................................................................................... 245 xviii CHAPTER 1 HISTORICAL, CURRENT, AND FUTURE PERSPECTIVES FOR CONTROLLED ENVIRONMENT HYDROPONIC FOOD CROP PRODUCTION IN THE UNITED STATES 1 Historical, Current, and Future Perspectives for Controlled Environment Hydroponic Food Crop Production in the United States Kellie J. Walters1, Bridget K. Behe1, Christopher J. Currey2, and Roberto G. Lopez1,3 1Department of Horticulture, Michigan State University, 1066 Bogue Street, East Lansing, MI 48824 2Department of Horticulture, Iowa State University, 008 Horticulture Hall, Ames, IA 50011 3To whom correspondence should be addressed; e-mail address: rglopez@msu.edu Walters, K.J., B.K. Behe, C.J. Currey, and R.G. Lopez. 2020. Historical, current, and future perspectives for controlled environment hydroponic food crop production in the United States. HortScience, 55:758–767. Received for publication: 3 Feb. 2020 Accepted for publication: 17 Mar. 2020 Published: 29 April 2020 2 Abstract. Controlled environment (CE) food crop production has existed in the United States for many years, but recent improvements in technology and increasing production warranted a closer examination of the industry. Therefore, our objectives were to characterize historical trends in CE production, understand the current state of the U.S. hydroponics industry, and use historical and current trends to inform future perspectives. In the 1800s, CE food production emerged and increased in popularity until 1929. After 1929, when adjusted for inflation (AFI), CE food production stagnated and decreased until 1988. From 1988 to 2014, the wholesale value of CE food production increased from $64.2 million to $796.7 million AFI. With the recent increase in demand for locally grown food spurring an increase in CE production, both growers and researchers have been interested in using hydroponic CE technologies to improve production and quality. Therefore, we surveyed U.S. hydroponic food crop producers to identify current hydroponic production technology adoption and potential areas for research needs. Producers cited a wide range of technology utilization; more than half employed solely hydroponic production techniques, 56% monitored light intensity, and more than 80% monitored air temperature and nutrient solution pH and electrical conductivity. Additionally, the growing environments varied from greenhouses (64%), indoors in multilayer (31%) or single-layer (7%) facilities, to hoop houses or high tunnels (29%). Overall, producers reported managing the growing environment to improve crop flavor and the development of production strategies as the most beneficial research areas, with 90% stating their customers would pay more for crops with increased flavor. Lastly, taking historical data and current practices into account, perspectives on future hydroponic CE production are discussed. These include the importance of research on multiple environmental parameters instead of single parameters in isolation and the emphasis on 3 not only increasing productivity but improving crop quality including flavor, sensory attributes, and postharvest longevity. Keywords: controlled environment agriculture (CEA); cucumber; culinary herb; greenhouse; lettuce; light-emitting diodes (LEDs); pepper; plant factory; strawberry; tomato; vertical farming CE food crop production has fluctuated in productivity and evolved in technology over history. For example, early U.S. CE food production generally took place in glasshouses heated by “hot beds” of manure (Dalrymple, 1973), whereas currently food is produced not only in glasshouses but plastic-glazed houses and indoor facilities heated through many methods, most of which are not manure-based. By examining both historical trends and current practices, we can deduce possible future CE hydroponic production trends. Therefore, our goals were to 1) characterize the historical trends in CE production, 2) conduct a national survey with the objectives of identifying current CE hydroponic production technology adoption and the research needs and priorities of the U.S. hydroponics industry, and 3) use historical trends and current practices to inform future perspectives. Historical Perspectives of U.S. CE and Hydroponic Production Although CE vegetable production has been reported to have originated during the Roman Empire, it was not until the early 1800s that commercial CE food production emerged in the United States. By 1900, an estimated 1000 facilities were growing 40 hectares (ha) of winter vegetables in CE with ≈67% of crops grown in greenhouses and ≈33% grown in hotbeds and coldframes (Dalrymple, 1973; Galloway, 1900; Jensen and Collins, 1985). At this time, 4 wholesale and retail annual CE vegetable sales were estimated to be $2.3 and $4.5 million, respectively (Dalrymple, 1973; Galloway, 1900). By 1929, the first year the U.S. Department of Commerce [USDC; these data are now collected by the U.S. Department of Agriculture (USDA)] Census of Horticultural Specialties reported vegetables grown under protected culture, 520 ha of vegetables were produced with 43%, 33%, 18%, and 6% of the crop value consisting of tomato (Solanum lycopersicum), cucumber (Cucumis sativus), lettuce (Lactuca sativa), and other crops, respectively, and sales totaling $10.0 million [$136.9 million adjusted for inflation to 2014 (AFI); USDC – Bureau of the Census, 1930; Table I-1]. Although food was grown under protected culture in glass-glazed greenhouses, hydroponic production was limited (Jensen and Collins, 1985). Hydroponic production can generally be defined as growing plants without mineral soil, using “an inert medium such as gravel, sand, peat, vermiculite, pumice, perlite, coco coir, sawdust, rice hulls, or other substrates” instead and adding the nutrients necessary for growth (Resh, 2013). Although this definition includes soilless substrate growing systems typical to potted floriculture production, by convention, hydroponic production excludes this growing method (Gomez et al., 2019). A more common description used by the USDA Census of Horticultural Specialties and this study, defines hydroponic production as food crops grown “in nutrient solutions without soil” (USDA – National Agriculture Statistics Service, 2015). Interest in hydroponic production arose from issues with fertilization and soil as a substrate (Dalrymple, 1973). This led to the development of sand culture (Shive and Robbins, 1937), water culture (Gericke, 1933), and then subirrigation (Withrow and Biebel, 1937) hydroponic production systems in the late 1920s and 1930s. Commercial production-scale hydroponics in the United States began during World War II when the U.S. military used 5 hydroponics to produce fresh food on Pacific islands during the war (Jensen and Collins, 1985), and by 1972, commercial hydroponic production greenhouses emerged across the United States (Dalrymple, 1973). From 1929, protected vegetable cultivation stagnated or decreased until 1988, after which CE food production again increased in popularity. For example, in 1929, food crops worth $136.9 million AFI were produced under 5.2 million m2 (USDC – Bureau of the Census, 1930). By 1988, only $64.2 million AFI worth of vegetables grown under 1.2 million m2 remained, a 53% AFI decrease in sales and a 77% reduction in production area (USDC – Economics and Statistics Administration, 1991; Table I-1). The stagnation and decrease in CE food production was due to many factors, possibly including increased trade with Mexico; refrigeration and interstate infrastructure, making long-distance perishable food shipment feasible; and also the burgeoning floriculture industry, the value of which increased from $82 million in 1929 ($555 million AFI to 1988) to just under $2 billion in 1988 (USDC – Bureau of the Census, 1930; USDC – Economics and Statistics Administration, 1991). However, CE vegetable production has been increasing since 1988, and by 2014, $796.7 million worth of vegetables was produced in 8.7 million m2 of protected environments (USDA – National Agriculture Statistics Service, 2000, 2015). A seemingly large increase in greenhouse food production took place between 1988 and 1998, during which the number of operations nearly doubled from 581 to 1015 producers, and sales grew from $64.2 million AFI to $322.2 million AFI (402% AFI increase in total value; USDC – Economics and Statistics Administration, 1991; USDA – National Agriculture Statistics Service, 2000). However, an even greater increase in the number of greenhouses producing food occurred between 1998 and 2014, when the number of operations increased by 6 148% (1506 additional operations) and sales increased from $322.2 million AFI to $797.7 million (148% AFI increase). Over those 16 years, large increases in cucumber ($60.0 million AFI increase, 339%), fresh cut herbs ($26.1 million AFI increase, 158%), lettuce ($42.0 million AFI increase, 311%), strawberry (Fragaria ×ananassa; $847,000 AFI increase, 1193%), and tomato ($230.5 million AFI increase, 135%) sales occurred, while pepper (Capsicum annuum) remained relatively steady ($13.7 million AFI decrease, 19%; USDA – National Agriculture Statistics Service, 2000, 2015; Table I-1). When comparing 1929 to 2014, both Census of Horticultural Specialties reports cited cucumber, lettuce, and tomato as the most commonly produced CE crops. Additionally, the 2014 report also cited fresh cut culinary herbs, pepper, and strawberry as top CE crops (USDC – Bureau of the Census, 1930; USDA – National Agriculture Statistics Service, 2015; Table I-1). The portion of protected culture crops grown hydroponically was first reported in 2009, with 73% of CE vegetable crops produced in protected culture grown hydroponically. Tomato was the highest value food crop grown in CE, with sales totaling $355 million AFI, and 89% of those fruits were produced hydroponically (USDA – National Agriculture Statistics Service, 2010; Table I-1). Similarly, 92% of the cucumber crop produced in CE was grown hydroponically, while only 4% and 3% of pepper and strawberry crops, respectively, grown in CE were produced hydroponically in 2009 (USDA – National Agriculture Statistics Service, 2010). In 2014, the top hydroponically produced CE crops were as follows: cucumber (30.0 million kg), fresh cut herbs (3.4 million kg), lettuce (7.0 million kg), and tomato (75.1 million kg) with 91%, 21%, 70%, and 86% of cucumber, fresh cut herbs, lettuce, and tomato crops, respectively, grown in protected culture produced hydroponically (USDA – National Agriculture Statistics Service, 2015). 7 Although peer-reviewed aquaponics industry surveys (Love et al., 2014, 2015; Villarroel et al., 2016) and vegetable industry surveys, including a “State of the Vegetable Industry Survey” administered by American Vegetable Grower magazine (Gordon, 2016), have been conducted, these reports have focused primarily on aquaponics or outdoor production. A “State of Indoor Farming” survey was conducted by Agrilyst, a climate control system company, in 2017 reportinging data on facility type, crops produced, yield, profitability, technology, and growers’ future plans (Agrilyst, 2017). However, this survey did not identify specific production practices and technology adoption, nor did it identify the U.S. CE hydroponic industry’s research needs. This information is valuable for firms entering CE production or considering it for a near- term opportunity, in addition to educational programming efforts to inform future entrepreneurs. Therefore, a more comprehensive survey focusing on specific production practices and research priorities was needed to focus CE research and extension efforts on topics most beneficial to current and future producers. Current U.S. Hydroponic Crop Production Practices: A Survey Approach. To achieve our objectives, we developed a 23-question online survey for the hydroponics industry (Appendix B; https://osf.io/rvtem/?view_only=ff2affcc7eb442118c12cc4cf08bd78e). The questions were designed to gather information pertaining to 1) demographics and business operations, 2) young plant propagation; 3) hydroponic food production (transplant to harvest), and 4) research priorities. The survey consisted of four open-ended, 16 closed-ended multiple choice, and two constant sum questions. Additionally, we asked respondents to rate several topics relating to hydroponic production on a 4-point Likert scale in terms of their perceived 8 benefit to their operation. The scale ranged from 1 (not at all beneficial) to 4 (very beneficial). Means and standard deviations were calculated from the responses. With the survey content and methods approved by the Michigan State University committee on research involving human subjects (IRB x17-241e; Appendix C), the survey was created on the SurveyMonkey (San Mateo, CA) online platform to improve ease of access to survey participants. Requests to advertise the survey and recruit participants were sent to several widely read industry trade publications focused on greenhouse and CE crop production, including GrowerTalks, Greenhouse Product News, Greenhouse Grower, HortAmericas, HortiDaily, Inside Grower, Produce Grower, and Urban Ag News. Each publicized the survey by providing a link to it through their websites, e-newsletters, and blogs (Kuack, 2017). Research topic benefit was analyzed using Tukey’s honestly significant difference test with JMP (version 12.0.1; SAS Institute Inc., Cary, NC). Descriptive statistics were used to represent and compare all other data. Results and Discussion Producers and production area. From the first advertisement (4 Apr. 2017) to the closing date (21 June 2017), we obtained 42 useful responses from 19 states. Fifty-three percent of the respondents produced hydroponic food crops solely, and an additional 30% produced food in-ground and hydroponically. Whereas 10% reported producing hydroponic food crops as well as floriculture crops, 5% were switching from floriculture to hydroponic food production. One firm reported growing hydroponically for breeding and research. Although more than half of respondents only used hydroponic production techniques, 45% grew hydroponically in addition to floriculture or 9 in-ground food production. Increasing diversification potentially increases economic benefits including social capital formation, greater profitability, improved labor management, and improved economic resiliency (Boody et al., 2005; Mishra et al., 2004). Many respondents appear to be food-centric in production, which helps extension programming target specific, focused groups of producers in meetings and the trade press. The distribution of area dedicated to hydroponic production for the responding firms was wide (Fig. I-1). Only one firm reported a hydroponic production area of less than 46 m2, and 19% of the firms reported hydroponic area in production of 9290 m2 or more. Hydroponic area in production for the most frequently reported category was 93 to 464 m2, and the mean and median area were 2629 m2 and 697 m2, respectively. Firms produced crops in greenhouses (64%), indoors in multilayer (31%) or single-layer (7%) facilities, and hoop houses or high tunnels (29%; data not shown). Indoor production, used by 38% of the growers surveyed, creates opportunity for increasing environmental control compared with traditional greenhouse production (used by 64% of respondents; data not shown). For example, greenhouses primarily rely on variable solar radiation for the majority or all of the photosynthetically active radiation used by plants, while light (photoperiod, spectrum, and quantity) provided to plants indoors is more precisely controlled. Additionally, temperature in a greenhouse is affected by solar radiation, nighttime long-wave radiation loss, and reduced insulation compared with indoor production, resulting in greater external environmental influences on temperature compared with indoor production. Finally, humidity (or vapor pressure deficit) can be readily increased in both indoor production and greenhouses. Reducing humidity in the two environments requires different strategies depending on outdoor humidity and temperature, energy costs, supplemental carbon dioxide 10 (CO2) use, and the willingness to introduce outside pests to indoor facilities (Gomez et al., 2019). However, the greater ability to control the growing environment in indoor compared with greenhouse production results in higher capital investment and operating costs. Crops and production systems. Leaf lettuce (e.g., green leaf, red leaf) was produced by 58% of the respondents (Fig. I-2). Head lettuce (e.g., boston/bibb/buttercrunch and cos/romaine) was produced by nearly half of the firms, and fresh cut culinary herbs [e.g., basil (Ocimum spp.), cilantro (Coriandrum sativum), parsley (Petroselinum crispum), dill (Anethum graveolens), mint (Mentha spp.), rosemary (Rosmarinus officinalis), sage (Salvia officinalis), and thyme (Thymus vulgaris), etc.] were produced by 43% of respondents. Fourteen percent of firms produced microgreens, whereas other leafy greens [e.g., kale (Brassica oleracea), swiss chard (Beta vulgaris subsp. vulgaris), and spinach (Spinacia oleracea)] were grown by a third of the respondents. Five firms reported exclusive production of one of those categories of leafy greens (culinary herbs, head lettuce, leaf lettuce, or microgreens). Combined, leafy greens production accounted for 69% of the production based on the number of producers growing the crop weighted by the percentage of production (Fig. I-2). In 2014, the USDA Census of Horticultural Specialties only differentiated between the total weight of produce grown under protection and the weight of produce grown hydroponically. Data regarding the number of operations, area in production, and sales of hydroponically produced crops were not available. However, comparisons in trends can be made. Mirroring the 2014 USDA Census of Horticultural Specialties report stating more producers grew lettuce (763 producers) than fresh cut culinary herbs (524 producers) in CE, leaf and head lettuce were the most common hydroponically produced crops by survey respondents, followed by culinary 11 herbs. However, the value of fresh cut culinary herbs produced in CEs in the United States ($70.9 million) was greater than lettuce ($55.5 million; USDA – National Agriculture Statistics Service, 2015; Table I-1). The USDA reported tomato as the most commonly produced CE crop based on number of producers, weight, and value; however, only one-third of survey respondents reported producing tomatoes. Similarly, only 19% of the respondents produced cucumbers, whereas the USDA reported a similar number of operations producing cucumbers as lettuce (USDA, 2015; Table I-1, Fig. I-2). Fewer operations grew eggplant (Solanum melongena; 7%), fruit [e.g., strawberries, blueberries (Vaccinium spp.), and melon (Cucumis and Citrullus spp.), etc.; 17%], pepper (14%), and root crops [e.g., beets (Beta vulgaris), radishes (Raphanus sativus), and carrots (Daucus carota subsp. sativus), etc.; 10%]. Understanding crop diversity is critical for the improvement of extension materials and prioritization of research. Knowing that many firms do not have monocultures underscores the need to identify species-specific environment responses and cultural conditions to classify and group crops, thereby simplifying production and improving production efficiencies. Nutrient-film technique (NFT) was the most frequently used hydroponic system (48% of respondents, 36% of production area), followed by dutch or bato bucket (33% of respondents, 18% of production area), then raft or deep-flow technique (DFT; 25% of respondents, 14% of production area; Fig. I-3). The hydroponic production system used largely depends on the crop being produced. For example, dutch or bato bucket and slab or hanging gutter systems are more commonly used to produce strawberries and high-wire crops such as cucumbers, eggplant, peppers, and tomatoes, whereas NFT, raft or DFT, aeroponics, and ebb-and-flow systems are more commonly used for leafy greens including lettuce, culinary herbs, and microgreens (Jensen, 1997; Peet and Wells, 2005; Fig. I-3). 12 Nutrient and water management. Municipal water was used by 62% of the firms, whereas well water was used by 29% of respondents. Fewer firms used reverse-osmosis or deionized (14%), reclaimed (19%), or other water sources (19%), including plasma-activated, spring, fish hatchery, gray, rain, and surface or pond water (data not shown). For young plant production, water was applied as overhead irrigation (50%), daily single-event subirrigation (39%), constant subirrigation (29%), or other (18%) including drip, fog, hand (type not specified), and multiple-event-per-day subirrigation (data not shown). In recirculating systems, only 43% reported filtering the nutrient solution, using in-line filters (10%), bio filter, mesh filter, soil sock, ultraviolet, (5% each), or other methods (13%; Table I-2). Water source is a key component of hydroponic culture. Because water used for plant production can vary widely in pH, alkalinity, electrical conductivity (EC), concentrations and ratios of specific nutrients, other chemical species, and the potential for pathogen contamination, understanding differences in water quality between sources is essential. For example, Argo et al. (1997) analyzed 4306 water samples from greenhouse producers across the United Statesand found the pH ranged from 2.7 to 11.3, EC varied from <0.01 to 9.8 dS·m‒1, and alkalinity from CaCO3 ranged from 0 to 1120 mg·L‒1. When comparing hydroponic producer responses to a 2013 survey of ornamental plant growers, a higher percentage of ornamental producers (55%) used well water, whereas municipal water was used by a lower proportion of producers (27%; Hodges et al., 2015). Even though municipalities charge for water consumption, the majority of producers in our survey used municipal water. This may be due to the quality standards municipal water is held to by the U.S. Environmental Protection Agency (2009), which sets standards for more than 90 contaminants. As a result, water from municipalities is often 13 more consistent in quality than well or surface water because it is tested and is safe for both human consumption and food crops, including hydroponic, production (Shaw et al., 2015). Additionally, the tendency of CEA facilities to be located in or near urban areas where municipal water is readily available could contribute to this trend. Although only used by 14% of producers surveyed, reverse-osmosis or deionized water is one method of ensuring consistent water quality with low nutrient contamination; however, the cost of these systems may be prohibitive. Reclaimed water (used by 19% of respondents) may reduce the amount of municipal or groundwater needed to produce a crop, but like surface water, it may be susceptible to both nutrient and pathogen contamination (Hanning et al., 2009; Runia, 1993). Due to the wide variation in quality and nutrients already present in the water (Argo et al., 1997) and the potential for pathogen contamination (Hanning et al., 2009), understanding water source inputs gives growers, extension personnel, and researchers a helpful starting point for managing nutrients. Nutrient solution management is integral to successful hydroponic production. While the EC is a measurement of a solution’s conductivity reflecting the total amount of fertilizer ions, pH of the nutrient solution and the relative proportion and concentration of these ions determines the availability and uptake of specific nutrients by the plant. Both EC and pH can be measured with commonly used sensors (as reported by 85% of respondents; Fig. I-4) while more difficult to measure parameters, including specific nutrients and dissolved oxygen, were measured by only 54% and 28% of respondents, respectively. Additionally, 60% of respondents reported using sensors and/or automatic pumps to manage or monitor nutrient solution EC and temperature, whereas 57% and 21% used sensors and/or automatic pumps to manage and monitor pH and dissolved oxygen, respectively. 14 Researchers have reported different trends in plant growth and development, nutrient uptake, and quality in response to EC. For example, as EC increased from 1.4 to 3.0 dS·m‒1, lettuce fresh weight decreased when pH was adjusted daily, EC was not adjusted, and solutions were replaced every 2 weeks in a DFT system (Samarakoon et al., 2006). In contrast, as EC increased from 0 to 4.8 dS·m‒1, fresh weight of pakchoi (Brassica campestris L. ssp. chinensis) increased; however, plants were only irrigated three times per week, and the hydroponic production system type is unknown (Ding et al., 2018). Additionally, high-wire tomatoes grown with ECs ranging from 2.5 to 5.0 dS·m‒1 irrigated daily and NFT-grown culinary herbs with ECs from 0.5 to 4.0 dS·m‒1 and daily pH and EC adjustment had similar fresh weight among EC treatments (Currey et al., 2019; Walters and Currey, 2018; Wu and Kubota, 2008). The authors hypothesize that EC may not limit growth when adjusted continuously but may become a limiting factor when nutrient solutions are adjusted periodically in recirculating systems. A more consistent trend is observed between crops regarding nutrient uptake. For example, as EC increased, N concentration increased in pakchoi (Ding et al., 2018) and culinary herbs (Currey et al., 2019; Walters and Currey, 2018) and lycopene, fructose, glucose, and total soluble solids increased in tomato (Wu and Kubota, 2008). Respondents who used recirculating systems varied in their practices for adjusting nutrient solution EC: 43% added specific nutrients; 38% added a complete, balanced prepackaged mix; 8% did not adjust EC; and 12% took other management steps including mixing their own fertilizer and adding water (data not shown). Growers most commonly replaced nutrient solutions completely every 1 to 3 months (28%), whereas 20% never completely replaced it; the remainder replaced the solution at differing intervals varying from every day to every 3 to 6 months (Fig. I-5). The wide variation in methods to account for nutrient depletion 15 and solution replacement aligns with the producers reporting research on formulations (mean = 3.1 ± 1.0) and nutrient solution pH (2.8 ± 1.0) being slightly to very beneficial (Fig. I-6). Additionally, with the majority of producers surveyed monitoring and automatically adjusting nutrient concentrations, specific recommendations to aid growers in using automation to manage nutrients is needed. Research on crop-specific recommendations for specific mineral nutrient concentrations and ratios to maintain a balanced nutrient solution could improve crop quality, reduce excessive nutrient accumulation, reduce nutrient deficiencies, and conserve water and labor. Growing environment: Monitoring and control. The number of environmental parameters monitored during propagation was less than those monitored during finished production. For example, 32% and 45% of producers surveyed monitored light intensity and daily light integral (DLI) during propagation, respectively, while 56% and 49%, respectively, monitored these parameters during finished production (Fig. I-4). The use of supplemental lighting was more common during propagation (54%) than during finished production (45%), while the use of sole-source lighting was similar between propagation and finished production (21% and 20%, respectively; Fig. I-7). Even though supplemental lighting was more common during propagation than during finished production (Fig. I-7), light intensity and DLI were monitored by more producers during finished production (Fig. I-4). Using supplemental lighting to augment low sunlight intensities during propagation can increase yields. For example, McCall (1992) stated that increasing the supplemental photosynthetic photon flux density (PPFD) provided to tomatoes from 30 to 90 µmol·m‒2·s‒1 during transplant production not only increased plant height, leaf number, leaf area, and dry mass, but also yields for the first 16 weeks of production. Walters and Lopez 16 (2018) reported increasing the sole-source light intensity provided to basil seedlings during the first two weeks of growth from 100 to 600 µmol·m‒2·s‒1 PPFD resulted in an 80% increase in fresh cut yieldafter transplanting and finishing plants in a common greenhouse environment. During finished production, light intensity is a key driver of yield. For example, as DLI increased from 2 to 20 mol·m−2·d−1, basil, cilantro, dill, oregano (Origanum vulgare), thyme, parsley, mint, and sage fresh weight increased by 8.1 g (thyme) to 175.1 g (dill) (Litvin, 2019). Crop quality including postharvest life, appearance, and flavor is also influenced by light intensity. For example, cucumber postharvest shelf life was prolonged, and the fruit color was a more desirable deep green when light intensities during production were higher (Lin and Jolliffe, 1996), and basil favor compounds increased as the DLI increased from 5 to 25 mol·m−2·d−1 (Chang et al., 2008). By monitoring light intensity and DLI, producers can decide whether ambient light intensities are adequate or if supplemental lighting is necessary to improve production. Producers reported research on light quality (mean = 3.2 ± 1.0) and DLI (mean = 3.1 ± 1.1) as being moderately to very beneficial (Fig. I-6). Light quality is particularly important to the 20% of respondents who used sole-source lighting, as 100% of the light is provided by electrical lighting compared with supplemental lighting, where a larger proportion of light is provided by the sun (Poel and Runkle, 2017). Light quality can influence crop quality; in tomato, red light increases while far-red light decreases carotenoid accumulation (Alba et al., 2000). Additionally, increasing blue light increases the plant quality of a variety of crops, including increasing lycopene and β-carotene in tomato fruits (Gautier et al., 2004) and increasing the concentration of flavor compounds basil, dill, and parsley (Ichimura et al., 2009; Litvin et al., 2020). 17 Given choices of air, substrate, and nutrient solution temperature, respondents indicated that air temperature was the most commonly monitored environmental parameter; 74% of producers measured air temperature during propagation and 82% measured it during finished production (Fig. I-4). The rate of plant development is primarily driven by temperature (Heins et al., 1998). Because temperature can be a readily controlled environmental factor with a large impact on growth, development, and plant quality, it is commonly manipulated by growers. Target temperatures for crop production depend on many factors including the cardinal temperatures for different crop species, interactions with other environmental parameters, target finishing or harvest dates, desired size and quality, crop production stage, cost of cooling and heating, the ability to control the environment, time of year, and growing location. Models quantifying air temperature effects on cucumber (Slack and Hand, 1983), culinary herbs (Chang et al., 2005; Currey et al., 2016; Walters and Currey, 2019), lettuce (Scaife, 1973; Seginer et al., 1991), pepper (Nilwik, 1981), and tomato (Adams et al., 2001) have been published. In general, researchers have found that as temperature increases above the base temperature of a crop, the growth rate increases linearly until a species-dependent optimum temperature, above which the growth rate decreases. Some researchers have also investigated the interaction of temperature, light intensity, and plant age determining the optimal temperature may also be dependent on those factors (Nilwik, 1981; Pearce et al., 1993). However, more robust models including more data points across a broader temperature range and the interaction of temperature and other environmental parameters including light intensity, photoperiod, and CO2 concentration, as well as cultural parameters including nutrient concentration and proportions, are needed. Additionally, temperaturet extremes can reduce fruit-set of flowering food crops and should be taken into consideration (Erickson and Markhart, 2002). Given commercial producers’ ability to 18 monitor temperature and the availability of research models on which to base growing decisions, the producers surveyed identified research on temperature as the fourth-lowest research priority, being moderately beneficial (mean = 3.0 ± 0.8; Fig. I-6). However, production recipes, including the interaction of temperature and other environmental parameters, are deemed the second- highest research priority (mean = 3.3 ± 1.1; Fig. I-6). Thirty-six percent of producers surveyed monitored CO2 concentration during finished production, whereas 32% monitored CO2 during propagation. One-third of respondents injected CO2 to reach target concentrations ranging from 350 to more than 1500 ppm (Table I-3). During propagation and finished production, the same number of producers reported monitoring relative humidity or vapor pressure deficit (46% to 47%; Fig. I-4). Increasing CO2 concentration to species-specific saturation points increases yield of cucumber (Wittwer and Robb, 1964), lettuce (Knecht and O’Leary, 1983; Wittwer and Robb, 1964), pepper (Fierro et al., 1994), strawberry (Wang and Bunce, 2004), and tomato (Morgan, 1971; Wittwer and Robb, 1964). However, only 36% of growers monitored CO2 during finished production. CO2 management was deemed the lowest research benefit as growers ranked it as slightly to moderately beneficial (mean = 2.7 ± 1.1; Figs. I-4 and I-6). Crop quality. When asked if their customers would pay more for crops with increased flavor, 90% of respondents responded affirmatively, whereas 10% reported only their wholesale customers would pay more if the crop had improved nutrition or color (data not shown). Managing the growing environment to improve crop flavor was cited as the most beneficial research area (mean = 3.4 ± 0.7; Fig. I-6). Many environmental and cultural factors can influence the flavor of crops including light intensity (Chang et al., 2008), light quality (Alba et al., 2000; Gautier et al., 19 2004; Litvin et al., 2020; Weisshaar and Jenkins, 1998), air temperature (Chang et al., 2005), CO2 concentration (Wang and Bunce, 2004), and nutrition (Benard et al., 2009). However, increases in secondary metabolites do not necessarily result in improved flavor. For example, in a sensory panel, consumers deemed the flavor of basil grown at 23 °C under 400 or 600 µmol·m−2·s–1 PPFD too intense, whereas plants grown under 100 µmol·m−2·s–1 PPFD were not flavorful enough; however, plants grown under 200 µmol·m−2·s–1 PPFD had preferred characteristics (Walters et al., 2019). Similarly, increasing secondary metabolite production in brassicas can lead to increased health-promoting glucosinolates, but with the side effect of a bitter taste (Bell et al., 2018). With producers indicating their consumers would pay more for increased flavor, continued research to determine which type and intensity of flavor is preferred and by which consumer segments is essential to improving crop quality to increase prices. Research priorities. Given the range of production systems, cultural practices, and environmental conditions affecting hydroponic food crop production, coupled with the lack of science-based production recommendations, additional research is warranted. However, research priorities should reflect the needs of the commercial industry. The most important research priorities, on a scale of 1 to 4, as reported by growers were manipulating the growing environment to improve crop flavor (mean = 3.4 ± 0.7), production recipes (e.g., lighting, CO2, temperature, nutrients) (mean = 3.3 ± 1.1), light quality [e.g., supplying different wavelengths of light using light-emitting diodes (LEDs) (mean = 3.2 ± 1.0), food safety guidelines (mean = 3.2 ± 0.9), postharvest recommendations (3.1 ± 0.9), and energy-use and resource-use management (mean = 3.1 ± 1.0). The topic with the lowest mean score was CO2 management (mean = 2.3 ± 1.1; Fig. I-6). These 20 priorities, based on their perceived benefit to growers, should be taken into consideration when determining research priorities for hydroponic food crop production. Future Perspectives and Conclusions Although CE hydroponic production has been used to grow food crops for many years, the recent increase in CE food production is creating need for additional research to increase production efficiencies and profitability, improve yields and produce quality, and address production challenges. Through this producer survey, we are establishing baseline data on the variability in production type, technology adoption, and research needs of hydroponic food crop producers. Educators, extension specialists, and researchers can use these data to better understand the current state of the U.S. hydroponics industry, identify gaps in both knowledge and technology adoption, provide extension resources or research to aid in filling those gaps, and educate future industry members on current practices. On the basis of the survey results and the authors’ professional opinion, research should not only emphasize increasing productivity but should also examine crop quality, especially how to improve crop flavor, sensory attributes, and postharvest longevity. This will increase potential for CE producers to grow and market a premium crop, creating opportunities for increasing profitability (Bi et al., 2012). Additionally, much recent research has focused on lighting technology, including LEDs and the effects of light spectrum (Mitchell et al., 2015). However, growers rated creation of production recipes as equally important. Although more difficult to research than light spectra in isolation, research on how environmental parameters and cultural practices interact will likely lead to more effective production strategies, taking both yield and plant quality into account. However, as interactions become more complex, data interpretation, 21 analysis, and implementation will as well (Boote et al., 1996). Some hydroponic production models exist for CE produced lettuce such as the NiCoLet model predicting nitrate concentrations and growth (Seginer et al., 1998), a modified SUCROS87 model (Spitters et al., 1989) to simulate the effects of DLI, ADT, and plant density on lettuce (Both, 1995), and an evapotranspiration prediction model based on CO2 and DLI (Ciolkosz et al., 1998). Additionally, many researchers have focused on modeling to determine biomass production without considering crop quality (Marcelis et al., 1998), an aspect of production that is increasingly important (Sadílek, 2019). Finally, although hydroponic nutrient solution research has been conducted for years, it has become clear that, as Hoagland and Arnon (1950) stated: “There is no one composition of nutrient solution which is always superior to every other composition.” Increased efficiencies can be realized in nutrient management by investigating the interaction of specific nutrient concentrations and the proportions relative to each other to determine what is necessary for both optimal growth and quality of the range of species grown commercially (Ahn, 2019). Taken together, multiparameter research working toward optimizing environmental (light quality, quantity, temperature, etc.) and cultural (nutrient solution, cropping duration, etc.) parameters is integral to improving CE hydroponic production. The “optimization” of these parameters is dependent on production goals, which have traditionally focused on improving yield; however, as apparent from this research, many producers are now also focusing on improving crop quality. Therefore, the “optimization” of growing parameters should take both productivity and quality into account. 22 APPENDICES 23 APPENDIX A TABLES AND FIGURES 24 Table I-1. The number of operations, area under protection, kilograms (kg) produced, total sales, and sales adjusted for inflation of food crops grown under protection in the United States (U.S.) between 1929 and 2014 as reported by the U.S. Department of Commerce or the U.S. Department of Agriculture. Values do not include mushroom production. Crop Year Number of operations protection Area under (m2) kg produced Sales ($) 1,281,480 Hydroponic Total - x - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1929y 1949v 1959v 1970v 1979t 1988s 1998r 2009q 2014q 1998 2009 2014 1929 1949 1959 1970 1979 1988 1998 2009 2014 1988 1998 2009 2014 1998 2009 2014 1929 1949 1959 1970 682 126 71 66 153 133 201 343 733 192 323 524 1,141 330 293 233 193 100 129 338 763 78 165 265 534 26 76 130 1,384 711 770 688 Cucumber Herbs, fresh cut Lettuce Pepper Strawberry Tomato - 104,415 85,579 194,725 264,402 280,381 558,440 1,021,655 438,967 550,822 1,287,079 1,434,866 - 936,543 1,035,689 479,844 160,536 108,604 255,762 402,270 26,477 140,005 114,271 327,205 3,437 87,236 57,879 1,958,458 2,637,788 1,999,083 - 3,247,951w 1,277,176u 477,766u 515,881u 3,426,000w 8,912,000w 12,226,000w Sales adjusted for inflation (AFI) ($)z 44,429,690 12,447,996 3,853,694 3,192,403 11,733,000 18,018,000 17,697,000 77,650,000 44,865,000 -- -- -- 70,929,000 70,929,000 1,832,505 25,067,382 1,393,021 13,577,079 2,455,882 19,809,314 3,061,278 18,943,966 4,557,000 15,607,000 4,047,000 8,182,000 9,330,000 13,505,000 53,823,000 59,628,000 55,547,000 55,547,000 500,000 1,011,000 5,277,000 7,368,000 2,191,000 2,427,000 5,996,000 5,996,000 49,000 71,000 525,000 582,000 918,000 918,000 4,130,451 56,501,671 10,077,398 98,219,360 16,152,412 130,286,469 14,034,821 86,851,037 12,034,713 32,940,105 --p 11,040,710 30,028,586 77,650,000w 30,995,000 16,112,055 3,457,599 7,002,196 - - 28,848 -- - 5,942 -- - - - - 9,947,417 - - 826,536 3,493,523 - 177,173 320,009 - - - - 25 Table I-1 (cont’d). 1979 1988 1998 2009 2014 Other Total 734 414 715 1,148 1,889 883 125 81 98 128 118 193 345 851 - 768 - - 866 581 1,015 1,476 2,521 - - - - - - - - - - - - - - - - - - - - - 87,330,003 796,080 298,865 459,583 746,785 1,254,000 5,001,000 46,891,000 59,753,000 17,447,000 1,211,270 13,282,000 26,853,000 599,318 1,608,245 117,856,000 170,597,000 3,712,591 145,475,102 129,071,160 320,454,000 355,017,000 75,111,721 401,133,000 401,133,000 3,957,391 10,889,816 527,314 2,912,888 3,707,028 144,410 4,621,295 107,160 4,295,000 80,547 10,111,000 148,645 67,875,000 360,185 11,911,971 101,350,000 112,281,000 1,339,941 33,375,916 184,491,000 184,491,000 1,614,654 10,006,987 136,888,560 5,202,118 13,046,460 127,157,322 2,546,185 19,545,643 157,656,504 3,823,156 18,358,765 113,608,700 3,227,511 91,388,000 26,684,000 1,966,386 31,743,000 64,176,000 1,199,471 2,939,824 222,624,000 322,248,000 6,619,063 227,835,320 165,649,393 553,270,000 612,943,000 8,668,132 236,744,055 150,190,920 796,664,000 796,664,000 49,604,952 86,600,944 - - - - - - - - - - - - - - 1929 1949 1959 1970 1979 1988 1998 2009 2014 1929 1949 1959 1970 1979 1988 1998 2009 2014 z Values adjusted for inflation to 2014 using the consumer price index (CPI; USDL, 2019). y Vegetables grown under glass, in cold frames, or other structures. x Data not available. w Total sales by operations, not just wholesale. v Vegetables grown under glass. u Value based on wholesale prices. t Vegetables grown under protection. s Greenhouse produced vegetables. r Greenhouse produced food crops. q Food crops grown under protection. p Withheld to avoid disclosing data for individual operations. * This table does not include data from the Census of Agriculture in years when the Census of Horticultural specialties was not conducted because, though protected culture food production was reported, the reports include mushroom cultivation. Therefore, the numbers cannot be directly compared to the numbers reported in this table as mushrooms were excluded. (USDC, 1930; USDC, 1952; USDC, 1962; USDC, 1973; USDC 1982; USDC, 1991; USDA, 2000; USDA, 2010; USDA, 2015) 26 Table I-2. The types of nutrient solution filters used by respondents of a 2017 United States hydroponic grower survey. Filter type In-line Unspecified Bio Mesh Soil sock Ultra violet (UV) Cloth Diatomaceous earth Paper Particulate Sand Total respondents (n) z Some respondents utilized multiple filter types. Number 4 3 2 2 2 2 1 1 1 1 1 17z 27 Table I-3. The target carbon dioxide (CO2) concentration as reported by respondents to a 2017 survey of hydroponic growers in the United States. CO2 concentration (ppm) >350 400-1200 500 700 700 – 1000 800 1000 1100 1200 >1500 Unknown Total respondents (n) Number 1 1 1 1 1 1 2 1 2 1 2 14 28 Figure I-1. The area of production dedicated to hydroponics per producer as reported by respondents to a 2017 hydroponic grower survey in the United States. n = 42 29 70 60 50 40 30 20 10 0 Fruit Other Peppers Tomatoes Cucumbers Eggplant Root crops Leaf lettuce Microgreens Head lettuce Culinary herbs Other leafy greens Vegetable transplants Crop ) % ( t n e c r e P Figure I-2. The proportion of crops grown based on number of producers growing the crop weighted by the percent value of that crop compared to total (left), and the percentage of producers growing each crop (right), based on respondents of a 2017 United States hydroponic grower survey. n = 42 30 Ebb and flow Other Aeroponics Slab/hanging gutter Dutch/bato bucket Nutrient film technique (NFT) Raft/deep-flow technique (DFT) 50 40 30 20 10 ) % ( t n e c r e P 0 Hydroponic production system Figure I-3. The proportion of hydroponic production systems used by respondents of a 2017 United States hydroponic grower survey, based on space dedicated to each system (left), and the percentage of respondents utilizing each type of production system based on frequency (right). n = 42 31 Propagation Finished production ) % ( t n e c r e P 100 80 60 40 20 0 Other S m A LI) Light intensity Daily light integral (D S C) O) O ) in the air 2 perature perature perature Nutrient solution pH Relative humidity/vapor pressure deficit pecific nutrients in the nutrient solution Nutrient solution electrical conductivity (E ygen in the nutrient solution (D Air tem ubstrate tem Nutrient solution tem bient or injected carbon dioxide (C Dissolved ox Figure I-4. The environmental and cultural parameters monitored by hydroponic growers in the United States during propagation (n = 38) and finished production (n = 39) based on responses to a 2017 survey of hydroponic growers in the United States. 32 Figure I-5. The frequency of nutrient solution replacement as reported by respondents of a 2017 United States hydroponic grower survey. n = 40 33 ab abc a a-d a-d a-d a-d a-d a-d a-d cd d bcd s t n e d n o p s e r f o r e b m u N 40 30 20 10 0 ns LI) gral (D ht inte aily lig nt e m e g a n a ns atio ulatio n form olutio nt s utrie N d n e m m o c st re arv sth o P e D s e u ality elin uid ht q afety g Lig d s o o F e m e-us urc o s d re n e a y-us erg n E s e cip n re ctio Pro u d or v p fla e cro v pro et to im n e m g a n a ntal m e m n viro n E s C ute erature ar attrib p m n te ultiv olutio nt s utrie d n n strate, a Air, s b u nt e m e g a n a n g a p H n p olutio nt s utrie N atio n/pro ctio nt pro g pla u d e m xid n dio o arb C n u o Y Not at all beneficial Slightly beneficial Moderately beneficial Very beneficial Weighted mean 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 n a e m d e t h g i e W Figure I-6. The degree of benefit of research topics as reported by respondents to a 2017 survey of United States hydroponic growers. Means ± sd were calculated by assigning values to the responses, 1 = not at all beneficial, 2 = slightly beneficial, 3 = moderately beneficial, 4 = 34 Figure I-6 (cont’d). very beneficial. Research topics that share letters do not differ by Tukey’s honestly significant difference test at P ≤ 0.05. n = 36 35 ) % ( t n e c r e P 60 50 40 30 20 10 Propagation Finished production No 0 Yes, I use sole-source lighting Yes, I use supplemental lighting Figure I-7. The use of supplemental or sole-source lighting based on responses to a 2017 survey of hydroponic growers in the United States. n = 40 36 APPENDIX B HYDROPONIC PRODUCTION SURVEY 37 Hydroponic Production Survey Researchers at Michigan State University are conducting a confidential industry survey to gain a better understanding of soilless hydroponic food production practices and areas of research needs. Input about your operation and ideas will help ensure that our research can benefit hydroponic food growers. Except for your time, there are no risks or conflicts of interest associated with participation in this study and participation is voluntary. The survey should take approximately 10 minutes or less to complete. All responses will remain confidential. If you have any questions regarding this survey, please contact Dr. Roberto Lopez at rglopez@msu.edu or 517-614-9617. Section A: About your business 1. In which state is your business located (select primary location if multiple sites)? _____ State (please specify) ____________ Outside the United States (please specify) _______________ a. Hydroponic food production only b. Hydroponic food production and floriculture production c. Hydroponic food production and in-ground food production d. Switching from floriculture production to hydroponic food production e. Switching from in-ground food production to hydroponic production f. Other (please specify) _____________ 2. Which best describes your business: 3. How many square feet (or m2) is dedicated to hydroponics? a. 100,000 ft2 + (9,290 m2 +) b. 50,000 – 99,999 ft2 (4645 – 9289 m2) c. 30,000 – 49,999 ft2 (2787 – 4644 m2) d. 20,000 – 29,999 ft2 (1858 – 2786 m2) e. 10,000 – 19,999 ft2 (929 – 1857 m2) f. 5,000 – 9,999 ft2 (465 – 928 m2) g. 1,000 – 4,999 ft2 (93 – 464 m2) h. 500 – 999 ft2 (46 – 92 m2) i. less than 500 ft2 (46 m2) 4. What environment(s) do you use for hydroponic production? (check all that apply) ____ Greenhouse ____ Hoop house/high tunnel ____ Indoor (single-layer) ____ Indoor (multi-layer; e.g. vertical farm, shipping container, warehouse) ____ Outdoor ____ Other (please specify) _____________ 38 5. What growing system(s) do you use for hydroponic food production? (The total should equal 100%.) ______% Nutrient film technique (NFT) ______% Raft/deep-flow technique (DFT) ______% Dutch/Bato bucket ______% Slab/hanging gutter ______% Aeroponics ______% Ebb and flow ______% Other (please specify) ______________ 6. Based on value, what approximate percentages of each of the following hydroponically produced crops do you grow? (The total should equal 100%.) ______% Tomatoes ______% Peppers ______% Cucumbers ______% Eggplant ______% Fruit (strawberries, blueberries, melons) ______% Root crops (beets, radishes, carrots) ______% Culinary herbs (basil, cilantro, parsley, dill, mint, rosemary, sage, thyme) ______% Lettuce (green leaf, red leaf, cos/romaine) ______% Head lettuce (Bibb/Boston, and Buttercrunch) ______% Other leafy greens (kale, Swiss chard, spinach, etc.) ______% Microgreens ______% Vegetable transplants ______% Ornamental transplants ______% Other (please specify) ______________ 7. What is your water source(s)? Check all that apply. _____ Municipal _____ Well _____ Reverse-osmosis (RO) or deionized _____ Reclaimed water (please specify source) ___________ _____ Other (please specify) ___________ Section B. Young plant propagation 8. Do you propagate young plants? a. Yes b. No 9. What method do you use to grow your young plants? (check all that apply) ____ Overhead irrigation ____ Constant sub irrigation ____ Sub irrigation once per day Other (please specify) _____________ 39 10. What do you monitor during young plant propagation? (check all that apply) ____ Instantaneous light intensity ____ Daily light integral (DLI) ____ Air temperature ____ Substrate temperature ____ Nutrient solution temperature ____ Nutrient solution pH ____ Nutrient solution electrical conductivity (EC) ____ Ambient or injected carbon dioxide (CO2) in the air ____ Dissolved oxygen in the nutrient solution (DO) ____ Relative humidity ____ Other (please specify) _________ 11. Are high-intensity supplemental or sole-source lighting systems used during propagation? a. Yes, I use supplemental lighting b. Yes, I use sole-source lighting b. No 12. What are the desired air temperature set points during propagation? This would be a primary setting or for the largest crop. Crop: _____________, Day: ________, Night: _____________ Section C. Hydroponic food production (transplant to harvest) 13. During hydroponic food production (transplant to harvest), what do you monitor? (check all that apply) ____ Light intensity ____ Daily light integral (DLI) ____ Air temperature ____ Substrate temperature ____ Nutrient solution temperature ____ Nutrient solution pH ____ Nutrient solution electrical conductivity (EC) ____ Specific nutrients in the nutrient solution ____ Ambient or injected carbon dioxide (CO2) in the air ____ Dissolved oxygen in the nutrient solution (DO) ____ Other (please specify) _________ 14. Do you use sensors and automatic pumps to manage and maintain nutrient solution: Electrical conductivity (EC)? ____ yes ____ no pH? ____ yes ____ no Temperature? ____ yes ____ no Dissolved oxygen? ____ yes ____ no 40 15. How do you adjust the nutrient solution electrical conductivity (EC) to account for depletion of nutrients? 16. How often do you completely replace your nutrient solution in a recirculating culture system? a. Add specific individual nutrients b. Add a prepackaged mix c. I do not adjust EC d. Other (please specify) ________________ a. Every 1 to 2 weeks b. Every 3 to 4 weeks c. Every 1 to 3 months d. Every 3 to 6 months e. I do not completely replace my nutrient solution f. Other (please specify) ________________ g. Not applicable 17. Do you filter your nutrient solution? a. Yes (How? please specify) ___________________ b. No 18. Do you use high-intensity supplemental or sole-source lighting systems? a. Yes, I use supplemental lighting b. Yes, I use sole-source lighting b. No 19. Do you inject carbon dioxide (CO2)? a. No b. Yes [target air concentration (ppm)] _________________ 20. Do you use different air temperature set points for different crops? a. Yes b. No 21. What are the desired air temperature set points for transplant to harvest production? If different temperature set points are used, please indicate temperatures for your largest-value crop. Crop: _____________, Day: ________, Night: _____________ 22. Will your customers pay more for crops with increased flavor? 41 Research area Plant response to daily light integral (DLI) Light quality (ex: supplying different wavelengths of light using LEDs) Air, substrate, and nutrient solution temperature Nutrient solution formulations Nutrient solution pH Carbon dioxide management Production (lighting, CO2, temperature, nutrients, etc.) Young plant production/ propagation Energy-use and resource-use management Cultivar attributes Manipulating the growing environment to improve crop flavor Postharvest recommendations Food safety guidelines Not at all beneficial Very beneficial Section D. Research focus 23. In your opinion, how much would research on the following hydroponic topics benefit your operation? Value to your business Moderately beneficial Slightly beneficial 24. Do you have any additional thoughts you’d like to share regarding hydroponic production? ________________________________ 42 APPENDIX C INTERNAL REVIEW BOARD APPROVAL 43 44 LITERATURE CITED 45 LITERATURE CITED Adams, S.R., K.E. Cockshull, and C.R.J. Cave. 2001. Effect of temperature on the growth and development of tomato fruits. Ann. Bot. 88:869–877. Agrilyst. 2017. State of indoor farming. 3 Mar. 2019. https://www.agrilyst.com/stateofindoorfarming2017/. Ahn, T. 2019. Establishment of theoretical models for the interpretation of nutrient variations in electrical conductivity-based closed-loop soilless culture systems. Seoul Natl. Univ., Seoul, South Korea, PhD Diss. Alba R., N.M. Cordonnier-Pratt, and L.H. Pratt. 2000. Fruit-localized phytochromes regulate lycopene accumulation independently of ethylene production in tomato. Plant Physiol. 123:363–370. Argo, W.R., J.A. Biernbaum, and D.D Warncke. 1997. Geographical characterization of greenhouse irrigation water. HortTechnology 7:49–55. Bell, L., O.O. Oloyede, S. Lignou, C. Wagstaff, and L. Methven. 2018. Taste and flavor perceptions of glucosinolates, isothiocyanates, and related compounds. Mol. Nutr. Food Res. 62(18):1700990. Benard, C., H. Gautier, F. Bourgaud, D. Grasselly, B. Navez, C. Caris-Veyrat, M. Weiss, and M. Genard. 2009. Effects of low nitrogen supply on tomato (Solanum lycopersicum) fruit yield and quality with special emphasis on sugars, acids, ascorbate, carotenoids, and phenolic compounds. J. Agr. Food Chem. 57:4112–4123. Bi, X., L. House, Z. Gao, and F. Gmitter. 2012. Sensory evaluation and experimental auctions: measuring willingness to pay for specific sensory attributes. Amer. J. Agr. Economics 94:562–568. Boody, G., B. Vondracek, D.A. Andow, M. Krinke, J. Westra, J. Zimmerman, and P. Welle. 2005. Multifunctional agriculture in the United States. BioScience 55:27–38. Boote, K.J., J.W. Jones, and N.B. Pickering. 1996. Potential uses and limitations of crop models. Agronomy J. 88:704–716. Both, A.J., 1995. Dynamic simulation of supplemental lighting for greenhouse hydroponic lettuce production. Cornell Univ., Ithaca, PhD Diss. Chang, X., P. Alderson, and C. Wright. 2005. Effect of temperature integration on the growth and volatile oil content of basil (Ocimum basilicum L.). J. Hort. Sci. Biotech. 80:593– 598. 46 Chang, X., P.G. Alderson, and C.J. Wright. 2008. Solar irradiance level alters the growth of basil (Ocimum basilicum L.) and its content of volatile oils. Environ. Expt. Bot. 63:216–223. Ciolkosz, D.E., L.D. Albright, and A.J. Both. 1997. Characterizing evapotranspiration in a greenhouse lettuce crop. Acta Hort. 456:255–262. Currey, C.J., J.R. Smith, and K.J. Walters. 2016. Air temperature affects the growth and development of cilantro, dill, and parsley. HortScience 51:S219 (abstr.) Currey, C.J., K.J. Walters, and N.J. Flax. 2019. Nutrient solution strength does not interact with the daily light integral to affect hydroponic cilantro, dill, and parsley growth and tissue mineral nutrient concentrations. Agronomy 9:389. Dalrymple, D.G. 1973. Controlled environment agriculture: A global review of greenhouse food production. U.S. Dept. Agr. Econ. Res. Serv. No. 145622 13 Mar. 2019. https://ageconsearch.umn.edu/record/145622/files/faer89.pdf. Ding, X., Y. Jiang, H. Zhao, D. Guo, L. He, F. Liu, Q. Zhou, D. Nandwani, D. Hui, and J. Yu. 2018. Electrical conductivity of nutrient solution influenced photosynthesis, quality, and antioxidant enzyme activity of pakchoi (Brassica campestris L. ssp. Chinensis) in a hydroponic system. PloS One 13(8):e0202090. Environmental Protection Agency. 2009. National primary drinking water regulations. 1 July. 2019. https://www.epa.gov/sites/production/files/2016- 06/documents/npwdr_complete_table.pdf. Erickson, A.N. and A.H. Markhart. 2002. Flower developmental stage and organ sensitivity of bell pepper (Capsicum annuum L.) to elevated temperature. Plant Cell Environ. 25:123– 130. Fierro, A., A. Gosselin, and N. Tremblay. 1994. Supplemental carbon dioxide and light improved tomato and pepper seedling growth and yield. HortScience 29:152–154. Galloway, B.T. 1900. Progress of commercial growing of plants under glass. USDA. 13 Mar. 2019. https://naldc.nal.usda.gov/download/IND43620781/PDF. Gautier H., A. Rocci, M. Buret, D. Grasselly, Y. Dumas, and M. Causse. 2004. Effect of photoselective filters on the physical and chemical traits of vine-ripened tomato fruits. Can. J. Plant Sci. 85:439–446. Gericke, W.F. 1933. Fertilizing unit for growing plants in water. U.S. Patent 1,915,884. Gomez, C., C.J. Currey, R.W. Dickson, H. Kim, R. Hernandez, N.C. Sabeh, R.E. Raudales, R.G. Brumfield, A. Laury-Shaw, A.K. Wilke, R.G. Lopez, and S.E. Burnett. 2019. Controlled environment food production for urban agriculture. HortScience 54:1448–1458. 47 Gordon, R. 2016. State of the vegetable industry survey shows growers optimistic about 2016. Amer. Veg. Grower. 13 Mar. 2019. https://www.growingproduce.com/vegetables/state- of-the-vegetable-industry-survey-shows-growers-optimistic-about-2016/. Hanning, I.B., J.D. Nutt, and S.C. Ricke. 2009. Salmonellosis outbreaks in the United States due to fresh produce: Sources and potential intervention measures. Foodborne Pathogens Dis. 6:635–648. Heins, R.D., B. Liu, and E.S. Runkle. 1998. Regulation of crop growth and development based on environmental factors. Acta Hort. 513:17–28. Hoagland, D.R. and D.I. Arnon. 1950. The water-culture method for growing plants without soil. Circular. 2nd Ed. California Agr. Expt. Sta. 347. Hodges, A.W., H. Khachatryan, M.A. Palma, and C.R. Hall. 2015. Production and marketing practices and trade flows in the United States green industry in 2013. J. Environ. Hort. 33:125−136. Hoshi, T. and T. Kozai. 1984. Knowledge-based and hierarchically distributed online control system for greenhouse management. Acta Hort, 148:301–308. Ichimura, M., H. Watanabe, W. Amaki, and N. Yamazaki. 2009. Effects of light quality on the growth and essential oil content in sweet basil. Acta Hort. 907:91−94. Jensen, M.H. 1997. Hydroponics worldwide. Acta Hort. 481:719–730. Jensen, M.H. and W.L. Collins. 1985. Hydroponic vegetable production. Hort. Reviews 7:483– 558. Knecht, G.N. and J.W. O'Leary. 1983. The influence of carbon dioxide on the growth, pigment, protein, carbohydrate, and mineral status of lettuce. J. Plant Nutr. 6:301–312. Kuack, D. 2017. Industry survey seeks information on hydroponic food production. HortAmericas blog. 13 Mar. 2019. https://hortamericas.com/blog/news/industry-survey- seeks-information-on-hydroponic-food-production/. Lin, W.C. and P.A. Jolliffe. 1996. Light intensity and spectral quality affect fruit growth and shelf life of greenhouse-grown long English cucumber. J. Amer. Soc. Hort. Sci. 121:1168–1173. Litvin, A.G. 2019. Quantifying the effects of light quantity and quality on culinary herb physiology. Iowa State Univ. Ames, PhD Diss. Litvin, A.G., C.J. Currey, L.A. Wilson. 2020. Effects of supplemental light source on basil, dill, and parsley growth, morphology, aroma, and flavor. J. Amer. Soc. Hort. Sci. 758–767. 48 growth and flowering of annual bedding plants and vegetable transplants. HortScience 52:1221–1228. Resh, H.M. 2013. Hydroponic food production: A definitive guidebook for the advanced home gardener and the commercial hydroponic grower. 7th ed. CRC Press Taylor and Francis Group, Boca Raton, FL. Runia, W.T. 1993. A review of possibilities for disinfection of recirculation water from soilless cultures. Acta Hort. 382:221−229. Love, D.C., J.P. Fry, L. Genello, E.S Hill, J.A. Frederick, X. Li, and K. Semmens, 2014. An international survey of aquaponics practitioners. PloS One 9(7):e102662. Love, D.C., J.P. Fry, X. Li, E.S. Hill, L. Genello, K. Semmens, and R.E. Thompson. 2015. Commercial aquaponics production and profitability: Findings from an international survey. Aquaculture 435:67–74. Marcelis, L.F.M., E. Heuvelink, and J. Goudriaan. 1998. Modelling biomass production and yield of horticultural crops: A review. Scientia Hort. 74:83–111. McCall, D. 1992. Effect of supplementary light on tomato transplant growth, and the after- effects on yield. Scientia Hort. 51:65–70. Mishra, A.K., H.S. El-Osta, and C.L. Sandretto. 2004. Factors affecting farm enterprise diversification. Agr. Finance Rev. 64:151–166. Mitchell, C.A., M.P. Dzakovich, C. Gomez, R. Lopez, J.F. Burr, R. Hernández, C. Kubota, C.J. Currey, Q. Meng, E.S. Runkle, and C.M. Bourget. 2015. Light-emitting diodes in horticulture. Hort. Rev. 43:1–87. Morgan, J.V. 1971. The influence of supplementary illumination and CO2 enrichment on the growth, flowering and fruiting of the tomato. Acta Hort. 22:187–199. Nilwik, H.J.M. 1981. Growth analysis of sweet pepper (Capsicum annuum L.) 2. Interacting effects of irradiance, temperature and plant age in controlled conditions. Ann. Bot. 48:137–146. Pearce, B.D., R.I. Grange, and K. Hardwick. 1993. The growth of young tomato fruit. I. Effects of temperature and irradiance on fruit grown in controlled environments. J. Hort. Sci. 68:1–11. Peet, M.M. and G.W.H Welles. 2005. Greenhouse tomato production, p. 277–278 In: E. Heuvelink (Ed.) Tomatoes. CABI Publishing, Wallingford, UK. Poel, B.R. and E.S. Runkle. 2017. Spectral effects of supplemental greenhouse radiation on 49 Sadílek, T. 2019. Perception of food quality by consumers: Literature review. European Res. Studies 22(1):57–67. Samarakoon, U.C., P.A. Weerasinghe, and W.A.P. Weerakkody. 2006. Effect of electrical conductivity (EC) of the nutrient solution on nutrient uptake, growth and yield of leaf lettuce (Lactuca sativa L.) in stationary culture. Trop. Agr. Res. 18:13–21. Scaife, M.A. 1973. The early relative growth rates of six lettuce cultivars as affected by temperature. Ann. Appl. Biol. 74:119−128. Seginer, I., G. Shina, L.D. Albright, and L.S. Marsh. 1991. Optimal temperature setpoints for greenhouse lettuce. J. Agr. Eng. Res. 49:209−226. Seginer, I., G.V. Straten, and F. Buwalda. 1998. Nitrate concentration in greenhouse lettuce: A modeling study. Acta Hort. 456:189−198. Shaw, A., C.J. Currey, and M.R. Evans. 2015. Focus on soilless and hydroponic systems. Greenhouse Grower 33(5):48−52. Shive, J.W. and W.R. Robbins. 1937. Methods of growing plants in solution and sand cultures. New Jersey Agr. Expt. Sta. Bul. 636. Slack, G. and D.W. Hand. 1983. The effect of day and night temperatures on the growth, development and yield of glasshouse cucumbers. J. Hort. Sci. 58:567–573. Spitters, C.J.T., H. Van Keulen, and D.W.G. Van Kraalingen. 1989. A simple and universal crop growth simulator: SUCROS87, p. 147–181. In: R. Rabbinge, S.A. Ward, H.H. van Laar (eds.). Simulation and systems management in crop protection. Ctr. Agricultural Publishing Documentation, Wageningen, Netherlands. Villarroel, M., R. Junge, T. Komives, B. König, I. Plaza, A. Bittsánszky, and A. Joly. 2016. Survey of aquaponics in Europe. Water 8:468. Walters, K.J. and C.J. Currey. 2018. Effects of nutrient solution concentration and daily light integral on growth and nutrient concentration of several basil species in hydroponic production. HortScience. 53:1319−1325. Walters, K.J. and C.J. Currey. 2019. Growth and development of basil species in response to temperature. HortScience. 54:1915−1920. Walters, K.J. and R.G. Lopez. 2018. Quantifying the influence of light intensity and CO2 concentration during sweet basil seedling production on subsequent growth, development, and volatile content. HortScience. 53(9):S89. (abstr.) 50 Walters, K.J., B.K. Behe, and R.G. Lopez. 2019. Consumer sensory preferences in response to manipulating fresh basil flavor through controlled environment light, temperature, and carbon dioxide management. HortScience. 54(9):S90. (abstr.) Wang, S.Y. and J.A. Bunce. 2004. Elevated carbon dioxide affects fruit flavor in field‐grown strawberries (Fragaria ×ananassa Duch). J. Sci. Food Agr. 84:1464–1468. Weisshaar, B. and G.I. Jenkins. 1998. Phenylpropanoid biosynthesis and its regulation. Current Opinion Plant Biol. 1:251–257. Withrow, R.B. and J.P. Biebel. 1937. Nutrient solution methods of greenhouse crop production. Purdue Univ. Ag. Expt. Station. Circ. 232. Wittwer, S.H. and W.M. Robb. 1964. Carbon dioxide enrichment of greenhouse atmospheres for food crop production. Econ. Bot. 18:34–56. Wu, M. and C. Kubota. 2008. Effects of high electrical conductivity of nutrient solution and its application timing on lycopene, chlorophyll and sugar concentrations of hydroponic tomatoes during ripening. Scientia Hort. 116(2):122-129. U.S. Department of Commerce (USDC) – Bureau of the Census. 1930. Fifteenth census of the United States, horticulture, statistics for the United States and for states, 1929 and 1930. U.S. Govt. Printing Office. 13 Mar. 2019. . U.S. Department of Commerce (USDC) – Bureau of the Census. 1952. 1949 U.S. Census of Horticultural Specialties. U.S. Govt. Printing Office. 13 Mar. 2019. . U.S. Department of Commerce (USDC) – Bureau of the Census. 1962. 1959 U.S. Census of Horticultural Specialties. 13 Mar. 2019. . U.S. Department of Commerce (USDC) – Social and Economics Statistics Administration. 1973. 1970 U.S. Census of Horticultural Specialties. 13 Mar. 2019. . U.S. Department of Commerce (USDC) – Bureau of the Census. 1982. 1979 Census of Horticultural Specialties. 13 Mar. 2019. . U.S. Department of Commerce (USDC) – Economics and Statistics Administration. 1991. 1988 Census of Horticultural Specialties. 13 Mar. 2019. . 51 U.S. Department of Agriculture (USDA) – National Agriculture Statistics Service. 2000. 1998 Census of Horticultural Specialties. 13 Mar. 2019. . U.S. Department of Agriculture (USDA) – National Agriculture Statistics Service. 2010. 2009 U.S. Census of Horticultural Specialties. 13 Mar. 2019. . U.S. Department of Agriculture (USDA) – National Agriculture Statistics Service. 2015. 2014 U.S. Census of Horticulture Specialties. 13 Mar. 2019. . U.S. Department of Agriculture (USDA) – Agricultural Marketing Service. 2016. National organic standards board hydroponic and aquaponics task force report. 13 Mar. 2019. . U.S. Department of Labor (USDL) – Bureau of Labor Statistics. 2019. CPI Inflation Calculator. 14 Dec. 2019. . 52 CHAPTER 2 MODELING GROWTH AND DEVELOPMENT OF HYDROPONICALLY GROWN DILL, PARSLEY, AND WATERCRESS IN RESPONSE TO PHOTOSYNTHETIC DAILY LIGHT INTEGRAL AND MEAN DAILY TEMPERATURE 53 Modeling growth and development of hydroponically grown dill, parsley, and watercress in response to photosynthetic daily light integral and mean daily temperature Kellie J. Walters and Roberto G. Lopez Acknowledgments We gratefully acknowledge Sean Tarr, Nate DuRussel, and Alex Renny for assistance, Erik Runkle and Jennifer Boldt for their thoughtful reviews, JR Peters for fertilizer, Grodan for substrate, LS Svensson for shade cloth, and Hydrofarm for hydroponic production systems. This work was supported by Michigan State University AgBioResearch (including Project GREEEN GR19-019), the USDA National Institute of Food and Agriculture (Hatch project MICL02472), and The Fred C. Gloeckner Foundation. The use of trade names in this publication does not imply endorsement by Michigan State University of products named nor criticism of similar ones not mentioned. 54 Abstract. In controlled environments, crop models that incorporate environmental factors can be developed to optimize growth and development as well as conduct cost and/or resource use benefit analyses. The overall objective of this study was to model growth and development of dill ‘Bouquet’ (Anethum graveolens), parsley ‘Giant of Italy’ (Petroselinum crispum), and watercress (Nasturtium officinale) in response to photosynthetic daily light integral (DLI) and mean daily temperature (MDT). Seeds sown in rockwool cubes were grown in a greenhouse at 23 ºC until transplant. Seedlings were transplanted into 0.9 × 1.8 m deep-flow hydroponic systems and grown in five greenhouse compartments with MDTs ranging from 9.7 to 27.2 ºC under 0%, 30%, or 50% shade cloth to create DLIs ranging from 6.2 to 16.9 mol·m-2·d-1. After two (watercress), three (dill), or four (parsley) weeks, maximum quantum yield of dark-adapted leaves (Fv/Fm), leaf number, branch number and length, height, and fresh and dry mass were recorded and dry matter concentration (DMC) was calculated. MDT and DLI interacted to influence dill fresh mass and height, and watercress Fv/Fm, height, and branch number while only MDT affected dill leaf number and watercress fresh mass and branch length. Besides DMC, parsley was influenced by MDT and not DLI. Increasing MDT from ≈10 to 22.4 ºC (parsley) or 27.2 ºC (dill and watercress), linearly or near-linearly increased fresh mass. For dill, increasing DLI decreased fresh mass when MDT was low (9.7 to 13.9 ºC) and increased fresh mass when MDT was high (18.4 to 27.2 ºC). DMC of dill, parsley, and watercress increased as MDT decreased or DLI increased, indicating a higher proportion of plant fresh mass is water at higher MDTs or lower DLIs. With these data we have created growth and development models for culinary herbs to aid in predicting responses to DLI and MDT. 55 Keywords: Controlled environment agriculture (CEA), culinary herb, environmental interaction, greenhouse, plant factory, vertical farming Abbreviations: DLI, daily light integral; DLIopt, optimal daily light integral; DMC, dry matter concentration; Fv/Fm, maximum quantum yield; MDT, mean daily temperature; MDTopt, optimal mean daily temperature; PPFD, photosynthetic photon flux density; Tb, base temperature; Tmax, maximum temperature; Topt, optimum temperature Introduction The fresh culinary herb market in the United States (U.S.) is at the introductory stage of its product life cycle, with growth of 10% to 14% per year from 2004 to 2014 (USAID, 2015). In 2014, 1.3 million m2 of controlled environment production yielded over 16.1 million kg of fresh- cut herbs (3.5 million kg from hydroponic systems) and total producer sales of $71 million (USDA, 2015). However, greenhouse herb growers face multiple challenges that impede their full production potential. From a recent survey of U.S. hydroponic growers, Walters et al. (2020) reported that the creation of “production recipes” taking multiple environmental parameters into account would be one of the most beneficial research topics to the industry. Plant growth, development, and quality are primarily influenced by light (radiant energy) and temperature (thermal energy). Photosynthesis and thus growth, biomass accumulation, crop quality, and yield are primarily influenced by the photosynthetic daily light integral (DLI; mol·m‒2·d‒1). In commercial greenhouse and indoor plant production, one main limitation to high rates of photosynthesis is low DLI. In the northern U.S., average outdoor DLI in December and January often falls below 10 mol·m‒2·d‒1, the threshold below which the growth of most 56 floriculture crops is acceptable (Faust and Logan, 2018), and greenhouse superstructure can further reduce DLI at the plant canopy by 35% to 70% (Both and Faust, 2017). It is well established that increasing the cumulative radiation intensity to a crop-specific optimum increases photosynthesis, biomass, and overall plant quality; thus, low DLIs during winter months can reduce plant quality (Faust et al., 2005; Litvin-Zabal, 2019). Recent research has determined that increasing DLI from 2 to 19 mol·m‒2·d‒1 [parsley ‘Giant of Italy’ (Petroselinum crispum)] or 20 mol·m‒2·d‒1 [dill ‘Fernleaf’ (Anethum graveolens)] increased fresh mass and harvestable yield (Litvin-Zabal, 2019). However, to increase fresh mass yield during low solar radiation conditions in a greenhouse or indoors, supplemental or sole-source lighting is required, respectively. Temperature is the primary determinant of developmental rates, including the progress to flower and leaf unfolding rate, but it also plays a role in growth and yield (Chermnykh and Kosobrukhov, 1987; Fraszczak and Knaflewski, 2009; Karlsson et al., 1991; Walters and Currey, 2019). More precisely, it is temperature integrated over time, typically a 24-h period, referred to as mean daily temperature (MDT). A temperature response curve can describe the relationship between development and MDT. Below the base temperature (Tb), development ceases. As MDT increases between Tb and the optimum temperature (Topt), development increases at a (near-) linear rate. At Topt, the rate of development is at its maximum. Temperature-dependent increases in photosynthesis, growth, and development in the linear range between Tb and Topt are largely due to increased enzymatic activity (Sage and Kubien, 2007). As MDT increases beyond Topt, the rate of growth and development decreases until the maximum temperature (Tmax), above which it ceases. 57 Many researchers have demonstrated that plant development in response to temperature is integrated over one day. Time to flower of African violet ‘Utah’ (Saintpaulia ionantha) was similar whether the day temperature was less than, the same as, or greater than the night temperature as long as MDT was the same (Faust and Heins, 1994). Similarly, Hurd and Graves (1984) determined that differences in day and night temperature did not influence time to flower or yield of tomato ‘Marathon’ (Solanum lycopersicum) as long as MDT was the same and the MDT was between Tb and Topt. Tomato ‘Counter’ truss and leaf number and early yields depended on MDT, regardless of night temperature, although later yields were higher when night temperature was higher at the same MDT (De Koning, 1988). Plant response to temperature can also be integrated over temporal durations longer than one day. De Koning (1990) determined that tomato ‘Counter’ can integrate temperatures over a 12 day period as long as the temperature fluctuation over the integration period was less than 6 °C, and Körner and Challa (2003) determined that chrysanthemum ‘Reagan Improved’ (Chrysanthemum morifolium) development can be integrated based on temperatures over a six day period. Chang et al. (2005) found that fresh mass of sweet basil ‘Sweet Genovese’ (Ocimum basilicum) did not differ when grown in six different temperature treatments, consisting of various combinations of one week each at 15, 25, or 30 °C but the same average temperature over a three week period. Radiation and temperature can interact to affect plant growth and development. Metabolic processes such as photosynthesis increase as temperatures increases to a process- specific Topt. Given this, both radiation and temperature impact photosynthesis. Additionally, dark respiration rates are affected by both night temperature and day radiation intensity. Respiration rate increases as temperature increases, reducing overall net carbon dioxide (CO2) 58 assimilation (Beinhart, 1962; Enoch and Hurd, 1977). Additionally, but to a smaller extent, increasing day radiation intensity increases dark respiration (Enoch and Hurd, 1977). In both acclimated and non-acclimated plants; photosynthetic Topt generally increases as radiation intensity increases (Chermnykh and Kosobrukhov, 1987; Enoch and Hurd, 1977). Increasing DLI from ~3 to ~13 mol·m‒2·d‒1 increased the photosynthetic Topt in both light- treatment acclimated and non-acclimated cucumber ‘Moskovsky Teplichnyi’ (Cucumis sativus) seedlings (Chermnykh and Kosobrukhov, 1987). In non-experimental condition acclimated carnation ‘Cerise Royalette’ (Dianthus caryophyllus), the Topt for whole-plant CO2 assimilation increased as radiation intensity increased (Enoch and Hurd, 1977). The CO2 assimilation Topt was between 5 and 10 °C when radiation intensity was 205 µmol·m–2·s–1 but increased to 27 °C when radiation intensity was 2,050 µmol·m–2·s–1. Models to predict growth and developmental parameters based on DLI and MDT in combination have been generated for economically important floriculture crops including petunia ‘Easy Wave Coral Reef’ and ‘Wave Purple’ (Petunia ×hybrida; Blanchard et al., 2011), celosia ‘Gloria Mix’ (Celosia argentea), impatiens ‘Accent Red’ (Impatiens walleriana; Pramuk and Runkle, 2005), salvia ‘Vista Red’ (Salvia splendens), marigold ‘Bonanza Yellow’ (Tagetes patula; Moccaldi and Runkle, 2007), cyclamen ‘Metis Scarlet Red’ (Cyclamen persicum; Oh et al., 2015), and pansy ‘Universal Violet’ (Viola ×wittrockiana; Adams et al., 1997a, b). Although MDT and DLI individually have been modeled in food crops mentioned previously, concurrent temperature and radiation-dependent modeling in food and agronomic crops has been more limited; models have been generated to predict the photosynthetic optimum of young cucumber ‘Moskovsky Teplichnyi’ (Chermnykh and Kosobrukhov, 1987) and white clover (Trifolium repens; Beinhart, 1962). To our knowledge, the influence of MDT and DLI on culinary herbs has 59 yet to be published. Understanding this interaction will allow us to develop more refined crop production models and provide recommendations to improve yields, improve grower profitability, and reduce excess energy costs. Although technology to manipulate the greenhouse temperature and DLI exists, its utility is limited when responses to both DLI and MDT are largely unknown for many crops. Hence, models to predict growth and development are integral to optimizing culinary herb production and conducting cost and/or resource use benefit analyses in response to environmental changes. Therefore, the overall objectives of this study were to determine the extent DLI and MDT influence the growth and development of culinary herbs and greens that can be readily produced in hydroponic production including dill, parsley, and watercress (Nasturtium officinale), and to create quantitative models to predict crop growth and development. Our hypothesis was that DLI would interact with MDT and the Topt of each species would increase as DLI increased. Materials and Methods Plant production. Seeds of dill ‘Bouquet’, parsley ‘Giant of Italy’, and watercress (Johnny’s Selected Seeds; Winslow, MA) were sown in trays containing stone-wool cubes (2.5 × 2.5 × 4 cm, AO plug; Grodan, Roermond, Netherlands) and placed in a greenhouse. Trays were irrigated overhead daily with reverse-osmosis water supplemented with 12N-1.8P-13.4K water-soluble fertilizer providing (mg·L–1) 100 nitrogen, 15 phosphorus, 112 potassium, 58 calcium, 17 magnesium, 2 sulfur, 1.4 iron, 0.5 zinc, 0.4 copper and manganese, and 0.1 boron and molybdenum (RO Hydro FeED; JR Peters, Inc., Allentown, PA) and magnesium sulfate (MgSO4) providing (mg·L–1) 15 magnesium and 20 sulfur. MDT (~23 °C) was measured by 60 aspirated and shielded 0.13-mm type E thermocouples (Omega Engineering, Norwalk, CT). High-pressure sodium lamps provided a photosynthetic photon flux density (PPFD) of ~80 µmol·m–2·s–1, as measured with a quantum sensor (LI-190R Quantum Sensor; LI-COR Biosciences, Lincoln, NE) every 15 s, and means were logged every hour by a CR-1000 datalogger (Campbell Scientific, Logan, UT) to maintain a 16-h photoperiod and a target DLI of 10 mol·m‒2·d‒1. On 25 Feb. 2018 (rep 1) and 1 Mar. 2019 (rep 2), three, four, and five weeks after sowing, watercress, dill, and parsley seedlings, respectively, were transplanted into 18-cm-deep by 0.9-m-wide by 1.8-m-long deep-flow hydroponic systems (270 L Active Aqua premium high- rise flood table; Hydrofarm, Petaluma, CA) in five connected glass-glazed greenhouse compartments with target constant MDTs of 10, 14, 18, 22, or 26 °C. Each greenhouse contained a hydroponic system under 0%, ~30%, and ~50% shade cloth (Solaro 3215 D O FB and Solaro 5220 D O; Ludvig Svensson, Kinna, Sweden) to create target DLIs of 12, 9, or 7 mol·m‒2·d‒1, respectively. Hydroponic net pots holding the seedlings were placed in 4-cm-diameter holes, 20-cm- apart, in 4-cm-thick extruded polystyrene floating on the nutrient solution. The nutrient solution consisted of reverse osmosis water supplemented with 12N-1.8P-13.4K water-soluble fertilizer (RO Hydro FeED; JR Peters, Inc.) and MgSO4 providing twice the concentrations reported previously. Electrical conductivity (EC) and pH were measured (HI991301 Portable Waterproof pH/EC/TDS Meter; Hanna Instruments, Woonsocket, RI) and adjusted to 1.56 mS·cm–1 and 6.0, respectively, by adding fertilizer, reverse osmosis water, potassium bicarbonate, or sulfuric acid. Air pumps (Active Aqua 70 L·min–1 commercial air pump; Hydrofarm) and air stones (Active Aqua air stone round 10 cm × 2.5 cm; Hydrofarm) were used to provide dissolved oxygen. 61 Exhaust fans, evaporative-pad cooling, radiant steam heating, and supplemental lighting were controlled by an environmental control system (Integro 725; Priva North America, Vineland Station, ON, Canada). The photoperiod was 16 h (0600 to 2200 HR), consisting of natural photoperiods (lat. 43º N) and day-extension lighting from high-pressure sodium lamps providing a supplemental PPFD of ~150 µmol·m–2·s–1 when the outdoor PPFD was low to maintain target DLIs. Shielded and aspirated 0.13-mm type E thermocouples (Omega Engineering) measured air temperature, infrared thermocouples (OS36-01-T-80F; Omega Engineering) measured leaf temperature of plants grown without shading, thermistors (ST-100; Apogee Instruments, Logan, UT) measured nutrient solution temperature, and quantum sensors (LI-190R Quantum Sensor; LI-COR Biosciences) placed at canopy height recorded PPFD in each treatment (reported as DLI, Table II-1). Every 15 s, a CR-1000 datalogger (Campbell Scientific) collected environmental data and hourly means were recorded. Growth data collection and analysis. The experiment was organized in a split-plot design with each of five MDTs in separate greenhouse sections and three DLI treatments in each section. The experiment was conducted twice over time. At transplant, dill, parsley, and watercress leaf number and fresh and dry mass, and watercress stem length were recorded. Watercress, dill, and parsley plants were harvested when one treatment reached individual marketable size, two, three, or four weeks after transplant. The most recent fully expanded leaf of five watercress and parsley plants in each treatment was dark acclimated for >15 min using manufacturer-supplied clips. Dark-acclimated leaves were exposed to 3,500 µmol·m–2·s–1 of red radiation (peak wavelength 650 nm) to saturate photosystem II, fluorescence was measured, and Fv/Fm was calculated and reported by a portable chlorophyll fluorescence meter (Handy Plant Efficiency Analyzer (PEA); Hanstech 62 Instruments Ltd. Norfolk, UK). The number of fully expanded dill and parsley leaves, number of watercress branches >2.5 cm, length of the longest watercress branch, plant height from the substrate surface to the tip of the tallest leaf, and fresh mass were recorded for 9 plants per treatment. Tissue was placed in a forced-air oven maintained at 75 ºC for at least 3 d, weighed, and dry mass was recorded. Dry matter concentration (DMC) was calculated as g dry mass per kg fresh mass. Seedling data were subtracted from harvest data for analysis. Analysis of variance was performed using JMP (version 12.0.1, SAS Institute Inc., Cary, NC); when interactions were not present, data were pooled. Linear, quadratic, and surface regression analyses were conducted using SigmaPlot (version 11.0, Systat Software Inc., San Jose, CA). Equations used to generate predictive models were based on 270 observations for each species. Results Dill. MDT and DLI interacted to influence dill fresh mass (Table II-2, Fig. II-1A). Within the observed ranges of MDT and DLI, MDT had a qualitatively larger effect than DLI. As MDT increased from 11.4 to 26.9 ºC when DLI was low (7.5 to 8.7 mol·m‒2·d‒1), fresh mass increased 11.7-fold (by 43.4 g). Increasing DLI had a negative effect on fresh mass when MDT was low (9.7 to 13.9 ºC) and a positive effect when MDT was high (18.4 to 27.2 ºC). For example, when grown at 9.7 or 13.9 ºC, increasing DLI from ~8.0 to ~15.5 mol·m‒2·d‒1, resulted in 25% (0.6 g) and 53% (6.6 g) less fresh mass, respectively. In contrast, increasing DLI from ~6 to ~15 mol·m‒ 2·d‒1, increased fresh mass 57% (19 g) at 27.2 ºC. Height at harvest followed a similar trend but increased over two-fold (>15.8 cm) as MDT increased from 9.7 to 21.6 ºC, then plateaued (Table II-2, Fig. II-1C). 63 The number of unfolded leaves increased linearly as MDT increased (Table II-2, Fig. II- 2F). Six more leaves unfolded when MDT increased from 13.6 to 27.2 ºC. DMC of dill increased as MDT decreased or DLI increased (Table II-2, Fig. II-3A‒B). As MDT increased from 9.7 to 27.2 ºC, DMC decreased by an average of 37% (53 g·kg‒1). As DLI increased by 4.3 to 8.4 mol·m‒2·d‒1, DMC increased by 12% (12 g·kg‒1). Parsley. Parsley DMC, fresh mass, maximum quantum yield of dark-adapted leaves (Fv/Fm), height, and leaf number were influenced by MDT, and DMC was also influenced by DLI. As MDT increased from 10.0 to 22.4 ºC, parsley fresh mass, height, and leaf number increased 13- fold (45 g), 3-fold (22.6 cm), and 1.3-fold (5 leaves), respectively (Table II-2, Figs. II-2A, C, and E). As MDT further increased from 22.4 to 27.1 ºC, fresh mass and height decreased by 29% (14 g) and 11% (3.2 cm), respectively, and leaf number did not increase. Fv/Fm was similar among plants grown at MDTs of 13.9 to 27.1 ºC. However, Fv/Fm declined at MDTs below 13.9 ºC (Fig. II-2G). Similar to dill, DMC of parsley increased as MDT decreased or DLI increased (Fig. II- 3A‒B). DMC was an average of 12% (19 g·kg‒1) lower as MDT increased from 10.0 to 27.1 ºC, and 11% (14 g·kg‒1) lower as DLI increased by 4.3 to 10.2 mol·m‒2·d‒1. Watercress. MDT but not DLI influenced watercress fresh mass and branch length (Table II-2, Figs. II-2B and D). As MDT increased from 10.1 to 27.2 ºC, fresh mass and branch length increased by 50-fold (20 g) and 11-fold (17 cm), respectively. MDT and DLI interacted to influence the watercress height at harvest. Plants were taller as MDT increased (Table II-2, Fig. II-1D). The effect of MDT was linear; increasing MDT from 64 11.7 to 26.8 ºC increased height 3-fold (15.1 cm) at a DLI of ~12 mol·m‒2·d‒1. The relationship between height and DLI was quadratic. For instance, at a MDT of 26.8 ºC, decreasing DLI from 12.2 to 9.9 mol·m‒2·d‒1 increased height 17% (3.4 cm) while further decreasing DLI to 8.8 mol·m‒2·d‒1 did not further change height. Watercress Fv/Fm was influenced by an interaction of MDT and DLI (Table II-2, Fig. II- 1B). MDT had a larger influence than DLI; Fv/Fm values were similar between plants grown from 13.9 to 27.2 ºC, but Fv/Fm values decreased as MDT decreased below 13.9 ºC. The effect of DLI was MDT-dependent. Decreasing DLI when MDT was high (13.9 to 27.2 ºC) had little effect on Fv/Fm (0.00 to 0.02 Fv/Fm increase) while decreasing DLI when MDT was low (10.1 to 13.7 ºC) resulted in greater Fv/Fm increases (0.04 to 0.05). Watercress branch number was also influenced by the interaction of MDT and DLI; increasing MDT increased branch number to a greater extent than increasing DLI (Table II-2, Fig. II-1E). Similar to dill and parsley, DMC of watercress was influenced individually by both MDT and DLI. DMC increased as MDT decreased or DLI increased (Table II-2, Figs. II-3A‒B). Discussion Photosynthetic DLI. The large role photosynthetic DLI plays in biomass accumulation and thus fresh yield has been well documented in food and ornamental crops (Beaman et al., 2009; Dou et al., 2018; Faust et al., 2005; Faust and Logan, 2018; Litvin-Zabal, 2019). In our experiment, the presence and extent of positive DLI effects were MDT dependent (Table II-2). Additionally, in some cases, DLI did not influence the growth nor developmental parameters measured. Though less frequently published, research has reported little to no differences in shoot biomass due to 65 increasing radiation intensity. For example, Liu and Su (2016) reported no difference in shoot dry biomass between Taxus grown under full sunlight, 40% to 60% full sunlight, and <10% full sunlight. Similarly, increasing the DLI from ~5 to ~20 mol·m‒2·d‒1 for gaura ‘Siskiyou Pink’ (Gaura lindheimeri) and ~6 to ~12 mol·m‒2·d‒1 for angelonia (Angelonia angustifolia ‘AngelMist White Cloud’) liners resulted in minimal to no dry mass increase (Currey et al., 2012; Fausey et al., 2005) While DLI did not affect parsley in our study (besides DMC), Litvin-Zabal (2019) reported that increasing DLI from 2 to 19 mol·m‒2·d‒1 increased fresh mass 4-fold (77 g) after four weeks at 22.7 °C MDT when three plants were grown per cell. Additionally, researchers have reported that increasing DLI from 7 to 18 mol·m‒2·d‒1 increased parsley ‘Giant of Italy’ fresh mass by 120% (13.3 g) four weeks after transplant (Currey et al., 2019). The lack of a DLI effect in our study may be due to low plant density reducing radiation interception competition or a long propagation time (five weeks) relative to finishing time (four weeks) where the treatments were applied. The influence of DLI on plant height and branch number at harvest without affecting fresh mass may be partially explained by DMC. An increase in DMC as DLI increases has been documented across many crops (Faust et al., 2005; Gent, 2014; Litvin-Zabal, 2019). Increasing DLI, to a certain extent, increases photosynthesis and carbon fixation, which increases carbohydrate accumulation and is reflected in a higher dry mass. In the case of watercress, although fresh mass was not affected by DLI, plants grown under higher DLIs had higher carbon fixation and higher dry masses, thus, higher DMC. Temperature. 66 Models to predict plant growth and development can be generated based on temperature response curves. Karlsson et al. (1991) developed a model to predict leaf unfolding rates of hibiscus ‘Brilliant Red’ and ‘Pink Versicolor’ (Hibiscus rosa-sinensis). Since the rate of development between Tb and Topt increases linearly as temperature increases, their models were most accurate when temperatures were within the linear range and plants were vegetative. In our study, parsley fresh mass increased as MDT increased from ~10.0 to 22.4 °C, while dill and watercress fresh mass increased from ~10.0 to 27.2 °C (Table II-2, Figs. II-1A, II-2A, and II- 2B). By fitting a quadratic curve for parsley, a Topt of 28 °C for fresh mass was calculated. The Topt will be >27.2 °C for dill and watercress fresh mass, since they had a linear response and supraoptimal temperatures were not attained. Our estimation is higher than Currey et al. (2016), who calculated parsley ‘Giant of Italy’ and dill ‘Fernleaf’ had a fresh mass Topt of 22.9 and 22.5 °C, respectively. We hypothesize the Topt differences between studies can be attributed to several factors including container vs. hydroponic production, different cultivars (dill), and a higher DLI (19.5 mol·m‒2·d‒1) in their study. Fraszczak and Knaflewski (2009) grew dill and parsley at MDTs of ~13, ~18, and ~23 °C, and DLIs of 2.9 and 3.8 mol·m‒2·d‒1. Fresh mass of dill was responsive to temperature and increased from 5.0 to 6.5 g as temperature increased, whereas parsley fresh mass was similar when grown at ~13 and ~23 °C. However, the DLIs utilized in this experiment were very low and unsuitable for commercial production. Though our regression models do not include data below 6.2 mol·m‒2·d‒1, their results concur with our findings for dill that when DLI was relatively low (7.5 to 8.7 mol·m‒2·d‒1), increasing MDT increased fresh mass. Additionally, that all three species had lower DMC with increased MDT may be explained by increased respiratory activity and reduced carbon fixation. 67 While watercress grows in cool streams across the U.S. (Howard and Lyon, 1952), high air temperatures in our study did not negatively influence growth since fresh mass increased as MDT increased from 10.0 to 27.2 °C (Table II-2, Fig II-2B). Likewise, Engelin-Eigles et al. (2006) found that increasing MDT from 10 to 20 °C increased fresh mass by ~50 g two weeks after transplant; however, they found that fresh mass was similar between 20 and 25 °C. When determining planting density, branch length can be taken into account if overlapping branches are undesirable due to harvesting and automation. Based on our results, to avoid overlapping branches when harvested after two weeks, watercress should be grown at least 4 cm apart when MDT is 13 °C and 40 cm apart when MDT is 26 °C, while DLI did not influence branch length (Table II-2). DLI and MDT interactions. The interaction of DLI and MDT was species- and parameter-dependent; it did not influence any parsley growth or developmental parameters measured but did influence all measured dill parameters except for DMC and leaf number. Additionally, watercress height, branch number, and Fv/Fm were influenced by the interaction of DLI and MDT, but fresh mass, branch length, and DMC were not (Figs. II-1‒3). In other studies, MDT and DLI interacted to influence flowering rate, flower number, height, and dry weight of salvia ‘Vista Red’, but only flowering rate of marigold ‘Bonanza Yellow’ and impatiens ‘Accent Red’, and height and flower number of celosia ‘Gloria Mix’ (Moccaldi and Runkle, 2007; Pramuk and Runkle, 2005). In general, increasing MDT has a greater positive effect on growth and photosynthesis when radiation intensity is higher (Beinhart, 1962; Chermnykh and Kosobrukhov, 1987). In our study, the influence of MDT on dill fresh mass was DLI dependent; increasing DLI had a small negative effect when MDT was low (9.7 to 13.9 ºC) but a positive effect when MDT was high 68 (18.4 to 27.2 ºC; Table II-2, Fig. II-1). These results agree with Litvin-Zabal (2019) who reported that at a MDT of 22.7 °C, increasing the DLI from 2 to 20 mol·m‒2·d‒1 increased dill fresh mass. However, based on our results, increasing dill fresh mass when MDT is 15 °C or below will require reducing the DLI to 11.4 mol·m‒2·d‒1 or below to achieve maximum fresh yield. Similar detrimental effects of increased radiation intensity at low temperatures have been documented; at 10 °C, increasing the DLI from 10 to 33 mol·m‒2·d‒1 negatively affected sorghum (Sorghum bicolor) photosynthetic rate and appearance (Taylor and Rowley, 1971). Additionally, in cucumber ‘Moskovsky Teplichnyi’ grown under varying radiation intensities and temperatures, researchers found the effect of temperature on CO2 assimilation was attenuated when the growing environment DLI was lower (~3 mol·m‒2·d‒1 compared to ~13 mol·m‒2·d‒1; Chermnykh and Kosobrukhov, 1987). Conversely, in celosia ‘Gloria Mix’, a greater increase in height at flowering occurred with increasing MDT when DLI was low, and the MDT-dependent increase in height was attenuated at higher DLIs (Pramuk and Runkle, 2005). When non-light treatment acclimated white clover was grown at 10 ºC, increasing radiation intensity from 170 to 900 µmol·m–2·s–1 did not influence CO2 uptake; however, increasing the growing temperature to 30 ºC increased CO2 uptake with increased radiation intensity (Beinhart, 1962). Similarly, in non-CO2, -MDT, or -radiation intensity acclimated carnations ‘Cerise Royalette’, when the CO2 concentration was 700 mg·L–1, increasing MDT had little effect on CO2 uptake when radiation intensity was 205 µmol·m–2·s–1 but a large effect when radiation intensity was 2,050 µmol·m–2·s–1 (Enoch and Hurd, 1977). Similarly, in the current study, increasing MDT from 11.4 to 27.2 °C increased fresh mass by 52.9 g when dill was grown under a DLI of ~15 mol·m‒2·d‒1, but only by 45.6 g when grown under ~8 mol·m‒2·d‒1 (Fig II- 1A). While some studies grow plants in a common environment (non-acclimated plants) to 69 generate radiation intensity and temperature photosynthetic response models, these models are only accurate for the environmental condition they are acclimated to (Chermnykh and Kosobrukhov, 1987). However, general trends can be drawn in parallel to acclimated responses. Optimal environmental parameters, Topt for example, can change depending on the growing environment (Chermnykh and Kosobrukhov, 1987; Enoch and Hurd, 1976). For example, in cucumber ‘Moskovsky Teplichnyi’, a 5 ºC increase in growing temperature resulted in a 1 ºC increase in photosynthetic (CO2 assimilation) Topt (Chermnykh and Kosobrukhov, 1987). Modeling. Identifying optimal conditions based on other environmental parameters is integral to improving production efficiencies to achieve the producer- and situation-dependent desired outcome, whether it is high biomass, compact plants, more leaves, a higher DMC, or parameters not included in this study including high phytonutrient concentrations and a longer postharvest life (Walters et al., 2020). For example, when DLI cannot be significantly altered but MDT can, increasing MDT is a useful strategy to increase fresh mass. However, beyond the optimal MDT, not only will yield be reduced, but excess resources will be spent heating the growing environment. To calculate the MDTopt or DLIopt based on DLI or MDT, respectively, supraoptimal MDTs or DLIs must be included in a study. In our study, supraoptimal conditions were observed for dill and watercress height at harvest, but largely not for other parameters measured. Therefore, based on surface regression models (Table II-2, Figs. II-1C‒D), we calculated the MDTopt for each parameter based on DLI by using !M!"opt = " + 2c(MDT) + #(DLI) = 0, and the DLIopt based on MDT using !!#$opt = $ + 2%(DLI) + #(MDT) = 0 (Figs. II-4A‒C). Similarly, MDTopt can be calculated based on DLI; the MDTopt for dill height (the MDT to produce the 70 tallest dill, an often undesirable characteristic) is 25 °C when DLI is 7 mol·m‒2·d‒1, but is 27 °C when DLI is 12 mol·m‒2·d‒1 (Fig. II-4B). Though the relationship between the MDT and DLIopt for dill height has a positive slope (Fig. II-4A), this relationship for watercress height has a negative slope, with the DLIopt decreasing as MDT increases (Fig. II-4C). Crop height is an important factor to consider when planning for hydroponic culinary herb production. In greenhouses, the distribution of supplemental lighting changes with the distance from the radiation source and plant height should be considered to ensure even radiation distribution for more uniform crops. In indoor vertical production, plant height is even more integral because space between layers should be as close as possible to maximize vertical efficiency. If plants are too tall, leaves can touch light fixtures, leading to non-uniform radiation distribution and potential leaf damage. Based on our models, when DLI is 10 mol·m‒2·d‒1 and MDT is 15 °C, dill, parsley, and watercress will be 18, 15, and 7 cm tall at harvest (3, 4, and 2 weeks, respectively); whereas if grown at 20 °C, the plants will be 30, 24, and 11 cm tall (Table II-2). Additionally, if plants are projected to be too tall, time to harvest can be reduced though yield will be impacted. Based on our models (Table II-2), increasing MDT from 15 to 25 °C will increase fresh mass by 30, 32, and 11 g for dill, parsley, and watercress, respectively, when DLI is 10 mol·m‒ 2·d‒1. If the DLI is 15 mol·m‒2·d‒1, the same increase in temperature will increase parsley and watercress fresh mass similarly, but increase dill fresh mass by 34 g. These equations can be used to more accurately predict yield and crop responses to changing MDT and DLI within the ranges evaluated. Models were generated using air temperature rather than plant temperature. Though plant responses are due to plant temperature rather than air temperature, air temperature can be used to 71 closely simulate plant temperature, especially when models take both radiation quantity and quality and air temperature into account. Additionally, air temperature is an environmental parameter more commonly monitored and adjusted by producers than leaf temperature (Cockshull, 1988; Walters et al., 2020). Models generated in this study were based on single plants grown under DLIs ranging from 6.2 to 16.9 mol·m‒2·d‒1 and MDTs from 9.2 to 27.2 °C for 2, 3, and 4 week production durations for watercress, dill, and parsley, respectively. Predicting growth and development outside of the DLI and MDT ranges used in these models could yield inaccurate predictions. Similarly, trends could be extrapolated (though less accurate) for differing plant densities and production durations; however, these extrapolations would only be qualitative and not quantitative. Conclusions These data allow us to begin modeling plant growth, development, and quality to predict plant responses and conduct cost-benefit analyses. Though technology to precisely manipulate and regulate the growing environment exists, its utility and application is limited when optimal growing conditions are not known. These data will serve as a foundation, allowing growers to calculate and implement the most advantageous growing environment by taking growth, development, and energy costs into account. 72 Author contributions KJW and RGL conceptualized and designed the study, KJW performed the experiments, conducted the data analysis, and prepared the manuscript, RGL obtained funding and revised the manuscript. Declaration of competing interest The authors declare that they have no conflict of interest. 73 APPENDIX 74 Table II-1. Average daily light integral (DLI; mol·m‒2·d‒1 ± SD) and mean daily air (MDT), leaf, and nutrient solution temperature over the two (watercress), three (dill), or four (parsley) week growing period for two replications over time. Data were collected every 15 s with means logged every hour. Rep. and transplant date DLI Temperature °C Leaf Solution Air Dill ‘Bouquet’ (Anethum graveolens) - - - - - - - z - 14.7 ± 3.6 11.4 ± 1.9 16.4 ± 1.1 11.5 ± 1.1 9.4 ± 1.3 11.6 ± 1.8 11.4 ± 1.9 7.5 ± 1.1 11.4 ± 1.9 9.5 ± 1.2 14.7 ± 3.3 13.6 ± 1.0 17.3 ± 2.3 14.3 ± 0.9 10.5 ± 1.6 13.6 ± 1.0 12.6 ± 0.8 7.0 ± 1.5 13.6 ± 1.0 11.9 ± 0.9 15.9 ± 3.6 18.0 ± 0.9 21.3 ± 1.4 16.8 ± 0.6 16.5 ± 1.9 12.2 ± 1.8 18.0 ± 0.9 8.8 ± 1.4 18.0 ± 0.9 15.8 ± 0.6 12.7 ± 2.0 21.6 ± 0.9 26.1 ± 1.0 19.2 ± 0.9 18.8 ± 0.8 11.3 ± 1.8 21.6 ± 0.9 8.3 ± 1.3 21.6 ± 0.9 18.5 ± 1.3 13.8 ± 3.4 26.9 ± 0.4 29.1 ± 0.7 23.6 ± 0.7 23.6 ± 0.5 10.6 ± 1.8 26.9 ± 0.4 8.7 ± 1.4 26.9 ± 0.4 22.0 ± 0.3 9.7 ± 1.1 13.0 ± 1.1 10.3 ± 1.0 15.6 ± 3.3 9.7 ± 1.1 11.5 ± 2.1 10.3 ± 0.9 8.2 ± 1.6 9.7 ± 1.1 15.3 ± 3.7 13.9 ± 0.5 17.1 ± 0.7 13.8 ± 0.4 11.2 ± 2.5 13.9 ± 0.5 13.5 ± 0.5 8.1 ± 1.7 13.9 ± 0.5 13.7 ± 0.1y 14.1 ± 3.5 18.4 ± 0.5 21.5 ± 0.5 17.3 ± 0.3 17.2 ± 0.3 10.3 ± 2.2 18.4 ± 0.5 8.1 ± 2.2 18.4 ± 0.5 15.7 ± 0.4 12.9 ± 3.8 22.5 ± 0.4 23.1 ± 0.7 19.9 ± 0.7 20.0 ± 0.4 9.8 ± 2.5 22.5 ± 0.4 6.5 ± 2.4 22.5 ± 0.4 19.8 ± 0.4 15.0 ± 4.0 27.2 ± 0.3 28.1 ± 0.4 24.1 ± 0.5 8.7 ± 2.3 27.2 ± 0.3 23.3 ± 0.4 23.1 ± 0.9 6.2 ± 2.3 27.2 ± 0.3 1 25 Feb. 2018 2 1 Mar. 2019 z data not collected - - - - - - - - - - - - - DLI Temperature °C - - - - - - - - Air Leaf Solution Parsley ‘Giant of Italy’ (Petroselinum crispum) 14.7 ± 3.3 10.8 ± 1.8 16.2 ± 1.0 11.2 ± 1.0 9.1 ± 1.2 11.6 ± 1.8 10.8 ± 1.8 7.6 ± 1.2 10.8 ± 1.8 9.0 ± 1.3 15.0 ± 3.2 13.3 ± 1.0 17.3 ± 2.3 14.1 ± 0.8 10.7 ± 1.7 13.3 ± 1.0 12.4 ± 0.8 6.9 ± 1.4 13.3 ± 1.0 11.6 ± 0.9 16.2 ± 3.3 17.9 ± 0.8 20.8 ± 1.4 16.7 ± 0.5 16.2 ± 1.6 12.0 ± 2.0 17.9 ± 0.8 8.6 ± 1.4 17.9 ± 0.8 15.7 ± 0.6 12.8 ± 1.9 21.3 ± 1.0 25.6 ± 1.2 19.0 ± 0.9 18.8 ± 0.8 10.7 ± 1.9 21.3 ± 1.0 8.5 ± 1.3 21.3 ± 1.0 18.5 ± 1.3 14.5 ± 3.6 26.9 ± 0.4 29.0 ± 0.6 23.7 ± 0.6 10.7 ± 2.0 26.9 ± 0.4 23.5 ± 0.4 9.1 ± 1.5 26.9 ± 0.4 21.9 ± 0.3 17.1 ± 4.2 10.0 ± 1.3 13.2 ± 1.4 10.4 ± 1.0 12.6 ± 2.9 10.0 ± 1.3 10.3 ± 1.0 8.7 ± 1.7 10.0 ± 1.3 17.1 ± 4.7 13.9 ± 0.8 17.2 ± 0.9 13.9 ± 0.6 11.6 ± 3.0 13.9 ± 0.8 13.6 ± 0.7 8.3 ± 1.6 13.9 ± 0.8 13.7 ± 0.1 y 16.1 ± 5.0 18.3 ± 0.5 21.5 ± 0.5 17.5 ± 0.4 17.3 ± 0.4 11.2 ± 2.9 18.3 ± 0.5 8.5 ± 3.0 18.3 ± 0.5 15.7 ± 0.4 14.4 ± 4.6 22.4 ± 0.4 23.5 ± 0.9 20.1 ± 0.8 20.1 ± 0.5 10.4 ± 3.2 22.4 ± 0.4 6.6 ± 2.3 22.4 ± 0.4 19.8 ± 0.4 16.9 ± 5.1 27.1 ± 0.4 28.2 ± 0.6 24.3 ± 0.5 9.8 ± 3.0 27.1 ± 0.4 23.5 ± 0.7 23.4 ± 1.0 6.7 ± 2.5 27.1 ± 0.4 - - - - - - - - - - - - - DLI Temperature °C Leaf Solution Air Watercress (Nasturtium officinale) - - - - - - - - 13.3 ± 2.5 11.7 ± 2.0 16.4 ± 1.2 11.7 ± 1.2 9.6 ± 1.4 11.1 ± 1.6 11.7 ± 2.0 7.3 ± 1.1 11.7 ± 2.0 9.8 ± 1.2 13.4 ± 1.8 13.7 ± 1.1 17.5 ± 2.1 14.3 ± 1.0 10.1 ± 1.4 13.7 ± 1.1 12.8 ± 0.8 6.7 ± 1.4 13.7 ± 1.1 12.0 ± 0.9 14.3 ± 1.9 18.0 ± 0.9 21.7 ± 1.3 16.9 ± 0.6 16.8 ± 2.0 11.6 ± 1.2 18.0 ± 0.9 8.5 ± 1.2 18.0 ± 0.9 16.0 ± 0.7 12.1 ± 1.6 21.4 ± 0.9 26.2 ± 1.0 19.2 ± 1.0 18.8 ± 0.9 10.7 ± 1.3 21.4 ± 0.9 8.1 ± 1.1 21.4 ± 0.9 18.5 ± 1.4 12.2 ± 1.5 26.8 ± 0.4 29.1 ± 0.7 23.6 ± 0.8 9.9 ± 1.2 26.8 ± 0.4 23.6 ± 0.6 8.8 ± 1.4 26.8 ± 0.4 22.0 ± 0.3 15.4 ± 3.3 10.1 ± 1.0 13.4 ± 1.0 10.9 ± 0.7 11.5 ± 2.2 10.1 ± 1.0 10.8 ± 0.6 8.3 ± 1.8 10.1 ± 1.0 15.3 ± 4.0 13.9 ± 0.6 17.2 ± 0.8 13.8 ± 0.5 11.6 ± 2.4 13.9 ± 0.6 13.7 ± 0.5 8.3 ± 1.8 13.9 ± 0.6 13.7 ± 0.1 y 14.7 ± 3.5 18.5 ± 0.5 21.5 ± 0.4 17.3 ± 0.3 17.4 ± 0.2 10.7 ± 2.2 18.5 ± 0.5 8.7 ± 1.8 18.5 ± 0.5 15.8 ± 0.4 13.9 ± 3.8 22.4 ± 0.4 22.9 ± 0.7 19.6 ± 0.7 19.9 ± 0.5 10.5 ± 2.3 22.4 ± 0.4 7.3 ± 2.2 22.4 ± 0.4 19.7 ± 0.5 15.4 ± 4.5 27.2 ± 0.3 28.1 ± 0.4 24.0 ± 0.5 9.4 ± 2.4 27.2 ± 0.3 23.2 ± 0.5 22.8 ± 0.9 6.9 ± 2.0 27.2 ± 0.3 - - - - - - - - - - - - - 75 Table II-1 (cont’d) y partial data reported 0 76 100 ) g ( s s a m h s e r f l l i D 80 60 40 20 0 16 14 12 DLI (mol·m-2·d-1) 10 8 A 30 25 20 M DT °C 15 10 6 ) m c ( t h g i e h l l i D 50 40 30 20 10 0 16 14 12 DLI (mol·m-2·d-1) 10 8 C 30 20 25 M DT °C 15 10 6 m F / v F s s e r c r e t a W 0.86 0.84 0.82 0.80 0.78 0.76 0.74 0.72 30 ) m c ( t h g i e h s s e r c r e t a W 25 20 15 10 5 0 ) . o n ( s e h c n a r b s s e r c r e t a W 16 14 12 10 8 6 4 2 0 B 25 20 MDT (°C) 15 10 6 8 12 10 D LI (m ol·m-2·d-1) 14 16 D 14 12 10 DLI (mol·m-2·d-1) 8 14 12 10 DLI (mol·m-2·d-1) 8 10 6 10 6 30 25 20 M DT (°C) 15 E 30 25 20 MDT (°C) 15 Figure II-1. Mean daily temperature (MDT) and daily light integral(DLI) effects on dill 77 Figure II-1 (cont’d). (Anethum graveolens) fresh mass (A) and height (C), watercress (Nasturtium officinale) maximum quantum yield of dark-adapted leaves (Fv/Fm; B), height (D), and branch number (E). Response surfaces represent model predictions. Coefficients for these models are presented in Table II 2. Models are each based on 270 individual measurements. 78 Table II-2. Regression analysis parameters, R2 or r2, and calculated base temperatures (Tb) or equation required to calculate Tb for dill (Anethum graveolens), parsley (Petroselinum crispum), and/or watercress (Nasturtium officinale) fresh mass, leaf number, height, dry matter concentration (DMC), maximum quantum yield of dark-adapted leaves (Fv/Fm), branch number, and branch length in response to mean daily temperature (MDT; °C) and daily light integral (DLI; mol·m‒2·d‒1). Unless noted, all models are in the form of: ! = y0 + a*MDT + b*DLI + c*MDT2 + d*DLI2 + e*MDT*DLI (c) MDT2 (d) DLI2 (e) MDT*DLI R2 or r2 Tb Parameter y0 (a) MDT (b) DLI Fresh mass (g) Leaf no. Height (cm) DMC (g·kg‒1) DMC (g·kg‒1) Fresh mass (g) Height (cm) Leaf no. Fv/Fm w DMC (g·kg‒1) -78.53 z (14.29) x -2.41 (0.94) -25.11 (9.47) 222.38 (25.50) 75.63 (13.63) -55.16 (26.30) -32.65 (6.79) -2.56 (6.70 E-1) -3.24 E-2 (1.79) 163.15 (9.75) 2.33 (4.08 E-1) 3.67 E-1 (4.91 E-2) 3.40 (5.58 E-1) -10.06 (2.90) 6.50 (3.02) 4.65 (7.82 E-1) 3.62 E-1 (3.50 E-2) 8.69 E-1 (1.77) -1.11 (5.10 E-1) 9.88 (2.22) 1.88 (1.36) 2.98 (1.20) 79 Dill ‘Bouquet’ -4.80 E-1 (8.63 E-2) -7.54 E-2 -1.28 E-1 (1.29 E-2) (5.32 E-2) 1.90 E-1 (7.72 E-2) Parsley ‘Giant of Italy’ -1.16 E-1 (8.03 E-2) -9.01 E-2 (2.08 E-2) 2.18 E-1 (2.09 E-1) 6.98 E-2 (3.63 E-2) 5.64 E-2 (2.23 E-2) 0.775 0.681 0.599 0.742 0.168 0.642 0.780 0.693 0.340 0.109 28.14 - 3.48*DLI + 0.156*DLI2 y 6.6 - - - 10.4 - 7.1 - - 1 Table II-2 (cont’d). DMC (g·kg‒1) 115.69 (8.09) 2.37 (6.76 E-1) Watercress 1.25 E-2 (6.7 E-3) -1.57 E-2 (4.2 E-3) -4. E-4 (5.9 E-5) 1.98 (6.70 E-1) 9.49 E-1 (4.22 E-1) -6.40 E-3 (2.0 E-3) 7.14 E-1 (2.93 E-1) 1.10 (1.84 E-1) 1.45 E-2 (2.60 E-3) 1.14 (7.89 E-2) 1.31 (8.88 E-1) -1.88 (6.12 E-1) -12.46 (4.50) -15.47 (2.84) 7.12 E-1 (3.04 E-2) -12.03 (1.52) -15.14 (1.71) 116.23 (11.78) 58.39 (12.93) Height (cm) Branch no. Fv/Fm Fresh mass (g) Branch length (cm) DMC (g·kg‒1) DMC (g·kg‒1) z Coefficients for model equations were used to generate Figs. II-1‒3. y Tb changes beased on DLI y Standard error (SE) w Exponential rise to a maximum model used: ! = y0 + a*(1-exp(-c*MDT)) -8.00 E-2 (2.84 E-2) -5.28 E-2 (1.79 E-2) 2.15 (1.16) -3.95 E-2 (1.16 E-2) 1.41 E-2 (7.3 E-3) 2. E-4 (1. E-4) 0.152 0.722 0.808 0.554 0.691 0.850 0.274 0.077 16.72 - 1.20*DLI + 0.0571*DLI2 y - - - - - 10.6 11.6 80 ) g ( s s a m h s e r f y e l s r a P ) m c ( t h g i e h y e l s r a P ) . o n ( s e v a e l y e l s r a P / m F v F y e l s r a P 70 60 50 40 30 20 10 0 40 35 30 25 20 15 10 5 0 10 8 6 4 2 0 0.85 0.80 0.75 0.70 0.65 0.60 0.55 0.00 ) g ( s s a m h s e r f s s e r c r e t a W ) m c ( h t g n e l h c n a r b s s e r c r e t a W ) . o n ( s e v a e l l l i D 30 25 20 15 10 5 0 25 20 15 10 5 0 8 6 4 2 0 A C E G D B D F 10 15 20 MDT (°C) 25 10 15 20 MDT (°C) 25 Figure II-2. Mean daily temperature (MDT) effects on parsley (Petroselinum crispum) fresh mass (A), height (C), leaf number (E), and maximum quantum yield of dark-adapted leaves 81 Figure II-2 (cont’d). (Fv/Fm; G), watercress (Nasturtium officinale) fresh mass (B) and branch length (D), and dill (Anethum graveolens) leaf number (F). Lines represent model predictions, with the coefficients for these models presented in Table II-2. Symbols (means ± SD) represent measured data (n = 27). 82 ) 1 - g k · g ( C M D 200 180 160 140 120 100 80 60 40 0 A B 10 15 20 MDT (°C) 25 6 8 10 12 14 DLI (mol·m-2·d-1) 16 18 Figure II-3. Mean daily temperature (MDT; A) and daily light integral (DLI; B) effects on dill (Anethum graveolens; ●), parsley (Petroselinum crispum; ), and watercress (Nasturtium officinale; ) dry matter concentration (DMC). Lines represent model predictions, with the coefficients for these models presented in Table II-2. Symbols (means ± SD) represent measured data used to generate the models (A, n = 27; B, n = 9). 83 ) 1 - d · 2 - m · l o m ( t p o I L D ) 1 - d · 2 - m · l o m ( t p o I L D 14 13 12 11 10 9 10 9 8 7 6 Dill height A Dill height B y = 7.34 + 0.22x y = 22.56 + 0.37x 10 8 Actual DLI (mol·m-2·d-1) 12 14 16 18 Watercress height C 6 y = 12.40 + 0.25x 10 20 15 Actual MDT (°C) 25 30 ) C ° ( t p o T D M 27.5 27.0 26.5 26.0 25.5 25.0 24.5 Figure II-4. Predicted optimal daily light integral (DLIopt; A) and optimal mean daily temperature (MDTopt; B) to achieve the tallest dill (Anethum graveolens) based on actual MDT and DLI, respectively. Predicted DLIopt (C) to achieve the tallest watercress (Nasturtium officinale) based on actual MDT. Equations were generated based on surface regression models (Figs. II-1C and D) with model coefficients reported in Table II-2. 84 LITERATURE CITED 85 LITERATURE CITED Adams, S.R., S. Pearson, and P. Hadley, 1997a. The effects of temperature, photoperiod and light integral on the time to flowering of pansy cv. Universal Violet (Viola × wittrockiana Gams.). Annals of Botany 80(1):107‒112. Adams, S.R., S. Pearson, and P. Hadley, 1997b. An analysis of the effects of temperature and light integral on the vegetative growth of pansy cv. Universal Violet (Viola× wittrockiana Gams.). Annals of Botany 79(3):219‒225. Beaman, A.R., R.J. Gladon, and J.A. Schrader, 2009. Sweet basil requires an irradiance of 500 µmol·m‒2·s‒ 1 for greatest edible biomass production. HortScience 44:64−67. Beinhart, G., 1962. Effects of temperature & light intensity on CO2 uptake, respiration, and growth of white clover. Plant Physiology 37(6):709−715. Blanchard, M.G., E.S. Runkle, and P.R. Fisher, 2011. Modeling plant morphology and development of petunia in response to temperature and photosynthetic daily light integral. Scientia Horticulturae, 129(2):313‒320. Both, A.J. and J.E. Faust, 2017. Light transmission. In: Light management in controlled environments. R. Lopez and E. Runkle (eds.). Meister Media Worldwide Willoughby OH. Chang, X., P. Alderson, and C. Wright, 2005. Effect of temperature integration on the growth and volatile oil content of basil (Ocimum basilicum L.). J. Hort. Sci. Biotechnol. 80:593‒ 598. Chermnykh, L.N. and A.A. Kosobrukhov, 1987. Effect of environmental factors on optimum temperature and photosynthetic intensity of plants adapted to various conditions. Biotronics 16:1‒11. Cockshull, K.E., 1988. The influence of energy conservation on crop productivity. Engineering Econ. Aspects Energy Saving Protected Cultivation 245:530‒536. Cockshull, K.E., D.W. Hand, and F.A. Langton, 1981. The effects of day and night temperature on flower initiation and development in chrysanthemum. Acta Hort. 125:101–110. Currey, C.J., J.R. Smith, and K.J. Walters, 2016. Air temperature affects the growth and development of cilantro, dill, and parsley. HortScience 51:S219 (abstr.). Currey, C.J., K.J. Walters, and N.J. Flax, 2019. Nutrient solution strength does not interact with the daily light integral to affect hydroponic cilantro, dill, and parsley growth and tissue mineral nutrient concentrations. Agronomy 9:389. 86 Currey, C.J., V.A. Hutchinson, and R.G. Lopez, 2012. Growth, morphology, and quality of rooted cuttings of several herbaceous annual bedding plants are influenced by photosynthetic daily light integral during root development. HortScience 47(1):25–30. De Koning, A.N.M., 1988. The effect of different day/night temperature regimes on growth, development and yield of glasshouse tomatoes. J. Hort. Sci. 63(3):465‒471. De Koning, A.N.M., 1990. Long-term temperature integration of tomato. Growth and development under alternating temperature regimes. Scientia Hort. 45:117–127. Dou, H., G. Niu, M. Gu, and J.G. Masabni, 2018. Responses of sweet basil to different daily light integrals in photosynthesis, morphology, yield, and nutritional quality. HortScience 53:496–503. Engelen-Eigles, G., G. Holden, J.D. Cohen, and G. Gardner, 2006. The effect of temperature, photoperiod, and light quality on gluconasturtiin concentration in watercress (Nasturtium officinale R. Br.). J. Agr. Food Chem. 54:328–334. Enoch, H.Z. and R.G. Hurd, 1977. Effect of light intensity, carbon dioxide concentration, and leaf temperature on gas exchange of spray carnation plants. J. Experimental Bot. 28(1):84–95. Fausey, B.A., R.D. Heins, and A.C. Cameron, 2005. Daily light integral affects flowering and quality of greenhouse-grown Achillea, Gaura, and Lavandula. HortScience, 40(1):114– 118. Faust, J.E. and R.D. Heins, 1994. Modeling inflorescence development of the African violet (Saintpaulia ionantha Wendl.). J. Amer. Soc. Hort. Sci. 119:727–734. Faust, J.E. and J. Logan, 2018. Daily light integral: A research review and high-resolution maps of the United States. HortScience 53:1250‒1257. Faust, J.E., V. Holcombe, N.C. Rajapakse, and D.R. Layne, 2005. The effect of daily light integral on bedding plant growth and flowering. HortScience 40:645–649. Frąszczak, B. and M. Knaflewski, 2009. Effect of light conditions and temperature on fresh yield of some spice plants grown in containers. Veg. Crops Res. Bul. 71:59–67. Gent, M.P., 2014. Effect of daily light integral on composition of hydroponic lettuce. HortScience 49:173–179. Hodges, A.G., C.R. Hall, M.A. Palma, and H. Khachatryan, 2015. Economic contributions of the green industry in the United States in 2013. HortTechnology 25:805‒814. Howard, H.W. and A.G. Lyon, 1952. Nasturtium officinale R. Br. (Rorippa Nasturtium- Aquaticum (L.) Hayek). J. Ecology 40:228–245. 87 Hurd, R.G. and C.J. Graves, 1983. The influence of different temperature patterns having the same integral on the earliness and yield of tomatoes. Acta Hort. 148:547‒554. Jung, J.H., M. Domijan, C. Klose, S. Biswas, D. Ezer, M. Gao, A.K. Khattak, M.S. Box, V. Charoensawan, S. Cortijo, and M. Kumar, 2016. Phytochromes function as thermosensors in Arabidopsis. Science 354(6314):886–889. Karlsson, M.G., R.D. Heins, J.O. Gerberick, and M.E. Hackmann, 1991. Temperature driven leaf unfolding rate in Hibiscus rosa-sinensis. Scientia Hort. 45:323–331. Körner, O. and H. Challa, 2003. Temperature integration and DIF in cut chrysanthemum. J. Hort. Sci. Biotechnol. 78:335–342. Legris, M., C. Nieto, R. Sellaro, S. Prat, and J.J. Casal, 2017. Perception and signaling of light and temperature cues in plants. Plant J. 90(4):683–697. Litvin-Zabal, A.G., 2019. Quantifying the effects of light quantity and quality on culinary herb physiology. Iowa State Univ. Ames, PhD Diss. Liu, W. and J. Su, 2016. Effects of light acclimation on shoot morphology, structure, and biomass allocation of two Taxus species in southwestern China. Scientific Reports. 6:35384. Lopez, R.G. and E.S. Runkle, 2004. The effect of temperature on leaf and flower development and flower longevity of Zygopetalum Redvale ‘Fire Kiss' orchid. HortScience 39:1630– 1634. Moccaldi, L.A. and E.S. Runkle, 2007. Modeling the effects of temperature and photosynthetic daily light integral on growth and flowering of Salvia splendens and Tagetes patula. J. Amer. Soc. Hort. Sci. 132:283–288. Oh, W., J. Kim, Y.H. Kim, I.J. Lee, and K.S. Kim, 2015. Shoot elongation and gibberellin contents in Cyclamen persicum are influenced by temperature and light intensity. Hort. Environ. Biotech. 56(6):762–768. Pramuk, L.A. and E.S. Runkle, 2005. Modeling growth and development of celosia and impatiens in response to temperature and photosynthetic daily light integral. J. Amer. Soc. Hort. Sci. 130(6):813–818. Sage, R.F. and D.S. Kubien, 2007. The temperature response of C3 and C4 photosynthesis. Plant Cell Environ. 30:1086–1106. Sams, C.E., 1999. Preharvest factors affecting postharvest texture. Postharvest Biol. Tech. 15(3):249–254. Taylor, A.O. and J.A. Rowley, 1971. Plants under climatic stress: I. Low temperature, high light effects on photosynthesis. Plant Physiology 47(5):713–718. 88 U.S. Agency for International Development -Acceso. 2014. The U.S. market for fresh herbs. Market Brief #20. http://pdf.usaid.gov/pdf_docs/PA00KP2C.pdf Accessed 23 Oct. 2018. U.S. Department of Agriculture (USDA) – National Agriculture Statistics Service. 2015. 2014 U.S. Census of Horticulture Specialties. 13 Mar. 2019. . Walters, K.J. and C.J. Currey, 2019. Growth and development of basil species in response to temperature. HortScience 54:1915−1920. Walters, K.J., B.K. Behe, C.J. Currey, and R.G. Lopez. 2020. Historical, current, and future perspectives for controlled environment hydroponic food crop production in the United States. HortScience, 55:758–767. 89 CHAPTER 3 THE INFLUENCE OF AVERAGE DAILY TEMPERATURE AND DAILY LIGHT INTEGRAL ON THE GROWTH, DEVELOPMENT, BIOMASS PARTITIONING, AND COLOR OF PURPLE BASIL, SAGE, SPEARMINT, AND SWEET BASIL 90 The influence of mean daily temperature and daily light integral on the growth, development, biomass partitioning, and color of purple basil, sage, spearmint, and sweet basil Kellie J. Walters, Sean Tarr, and Roberto G. Lopez This work was supported by Michigan State University AgBioResearch (including Project GREEEN GR19-019), the USDA National Institute of Food and Agriculture Hatch project MICL02472 and The Fred C. Gloeckner Foundation. We gratefully acknowledge Nate DuRussel and Alex Renny for assistance, Randolph Beaudry and Brad Rowe for equipment, JR Peters for fertilizer, Smithers-Oasis for substrate, and Hydrofarm for hydroponic production systems. The use of trade names in this publication does not imply endorsement by Michigan State University of products named nor criticism of similar ones not mentioned. 91 Abstract. Mean daily temperature (MDT) and daily light integral (DLI) can interact to influence growth and development of greenhouse crops and be utilized to improve culinary herb crop timing, maximize biomass production, and increase quality. Our objectives were to 1) determine the extent DLI and MDT influence growth and development of purple basil ‘Dark Opal’ (Ocimum basilicum), sage ‘Extrakta’ (Salvia officinalis), spearmint ‘Spanish’ (Mentha spicata), and sweet basil ‘Nufar’ (Ocimum basilicum); and 2) determine to what extent DLI and MDT influence purple basil color. Seedlings and liners were transplanted into one of five greenhouse compartments with ADT set points of 23, 26, 29, 32, or 35 °C. Each greenhouse contained three 0.9-m-wide by 1.8-m-long deep-flow hydroponic systems under 0%, 30%, or 50% shade cloth used to create target DLIs of 12, 9, or 7 mol·m‒2·d‒1, respectively. After 3 (mint and sweet basil), 4 (purple basil), or 5 (sage) weeks, growth and development and leaf color of purple basil was measured. MDT and DLI interacted to influence many growth and developmental parameters. Increasing MDT from ~23 to ~35 °C increased the branch number of all genera. Sweet basil branch number increased as DLI increased from 5.5 to 13.2 mol·m‒2·d‒1, but the effect of DLI was attenuated as MDT decreased. In contrast, increasing DLI from ~5-6 to ~18-19 mol·m‒2·d‒1 increased sage and spearmint branch number more when MDT was lower (~23 °C) compared to ~35 °C, and purple basil branch number was not influenced by DLI. The optimal MDT (MDTopt) for sage and spearmint fresh mass decreased from 27.5 to 23.5 °C and from 30.4 to 27.8 °C, respectively, as DLI increased from 6 to 18 mol·m‒2·d‒1, while sweet basil fresh mass MDTopt increased from 32.6 to 35.5 °C as DLI increased from 6 to 11 mol·m‒2·d‒1. Purple basil was greener [hue angle (h°) = 99° to 138°] when MDT was ~35 °C regardless of DLI, but when MDT was lower (~25 °C), basil was more purple (h° = 335°) at a DLI of 18.7 compared to 5.0 92 mol·m‒2·d‒1 (h° = 98°). Taken together, MDT and DLI can have a large impact on plant morphology, growth, development, and color. Models created in this study can serve as a grower decision-support tool to improve culinary herb yield and quality. Introduction Culinary herbs are commonly produced in controlled environments (CE) due to their high value, relatively short production duration, limited post-harvest life, and compact height making them conducive for both greenhouse and indoor vertical production. From 1998 to 2014, United States (U.S.) protected environemnt fresh cut culinary herb production increased in value by 58% (adjusted for inflation: $26 million increase) and 16.4 million kg were produced under 1.3 million m2 of protected production area (USDA, 2010; USDA, 2015). CE agriculture (CEA), including greenhouse and indoor production, creates the opportunity to precisely control the environment to potentially increase crop productivity and quality, provide a consistent year- round supply of locally grown food, and increase food safety and security; additionally, recent technological advances have improved the economic feasibility of CEA (Busch et al., 2020; Gomez et al., 2019; Kolodinsky et al., 2020). Though the ability to precisely control the growing environment exists, so does a knowledge gap in how to use the technology to its greatest potential. If known, increased production efficiencies including greater yields, increased energy use efficiency, and higher quality products could be realized. Both light (radiant energy) and temperature (thermal energy) play large roles in plant growth, development, and quality. While radiant energy is measured instantaneously, to a large extent, plants integrate the accumulation of radiant energy over the course of a 24-h period. Therefore, the daily light integral (DLI; mol·m‒2·d‒1) is a useful calculation to integrate 93 fluctuations of radiant energy in a greenhouse to meaningful numbers to predict plant responses over time (Korczynski et al., 2002). Photosynthesis, and thus biomass production, growth, and secondary metabolite production, are largely driven by radiation. Under low DLIs either indoors or during greenhouse production in the winter months can result in low-quality and -yielding crops (Korczynski et al., 2002). Increasing the DLI from 9.3 to 17.8 mol·m‒2·d‒1 during indoor sweet basil ‘Improved Genovese Compact’ (Ocimum basilicum) production not only increased fresh mass 78%, but also anthocyanin concentrations (Dou et al., 2018). Since increasing DLI can increase anthocyanin concentrations, a phenolic pigment contributing to blue, red, and purple coloration, we hypothesized that increasing DLI would not only improve biomass production, but also make purple basil (Ocimum basilicum) appear more purple and less green. Plant developmental rates, including leaf unfolding rate and progress towards flowering, are primarily influenced by temperature (Chermnykh and Kosobrukhov, 1987; Fraszczak and Knaflewski, 2009; Karlsson et al., 1991; Walters and Currey, 2019). However, temperature also plays a role in growth and yield. Similar to radiation, plant response to temperature integrated over one day is referred to as the mean daily temperature (MDT). The relationship between MDT and plant growth and development can be described by the temperature response curve. Below the species- and cultivar-specific base temperature (Tb), plant growth and development does not progress. As temperature increases above Tb, plant growth and development increases linearly or near-linearly until the optimum temperature (Topt) where the growth or developmental rate is the greatest. Increased enzymatic activity caused by increased temperature are largely responsible for the temperature-dependent increases in photosynthesis and growth in the linear range between Tb and Topt (Sage and Kubien, 2007). As MDT increases above Topt, growth and 94 development rates decrease until the maximum temperature (Tmax), beyond which plant development ceases. Several researchers have investigated the influence of temperature on culinary herbs. Chang et al. (2005) determined the Tb for sweet basil ‘Genovese’ growth was 10.9 °C. Similarly, Walters and Currey (2019) determined the Tb for growth of holy basil ‘holy’ (Ocimum tenuiflorum), sweet basil ‘Nufar’, lemon basil ‘Sweet Dani’, and lemon basil ‘Lime’ (Ocimum ´citriodorum), were 10.9, 11.3, 11.6, and 12.1°C, respectively. Chang et al. (2005) found that growth of sweet basil ‘Genovese’ increased as temperature increased from 15 to 25 °C, but plateaued as temperature further increased to 30 °C. Therefore, the Topt for sweet basil ‘Genovese’ was between 25 and 30 °C when the DLI was 20 to 22 mol·m‒2·d‒1. Walters and Currey (2019) determined the Topt was between 29 and 35 °C when the DLI was 19.5 mol·m‒2·d‒ 1. For basil, Tmax is a very high temperature not commonly reached in production. In another study, six species of sage besides Salvia officinalis (common sage) were grown for 36 d with 15- h daytime temperatures of 20, 25, 30, 35, or 40 °C and 9-h nights of 15 or 20 °C; maximum dry weight occurred at MDTs ranging from 24 to 37 °C and varied among species (Lasseigne et al., 2007). For common sage, fresh mass increased from 37 to 72 g and plant height increased from 9 to 18 cm as temperature increased from 18 to 27 °C (Mortensen, 2014). Additionally, metabolic processes including pigment production are influenced by MDT; as temperature increases, anthocyanin concentrations decrease, leading to less red/blue/purple plants (Lin-Wang et al., 2011; Mori et al., 2007; Movahed et al., 2016; Rehman et al., 2017). Temperature and radiation can interact to influence plant growth and development. Not only overt characteristics can be explained and modeled by a temperature response curve, including leaf unfolding rate and biomass accumulation, but also metabolic processes including 95 photosynthesis and respiration increase as temperature increases to a process-specific Topt. While examining the interaction of radiation and temperature on photosynthesis, researchers have determined that Topt generally increases as radiation intensity increases (Chermnykh and Kosobrukhov, 1987; Enoch and Hurd, 1977). For example, the carbon dioxide (CO2) assimilation Topt was 5 to 10 °C when the radiation intensity was low (205 µmol·m–2·s–1) but increased to 27 °C when the radiation intensity increased to 2,050 µmol·m–2·s–1 for non-light or - temperature acclimated carnation ‘Cerise Royalette’ (Dianthus caryophyllus; Enoch and Hurd, 1977). Similarly, the Topt of both light-acclimated and non-acclimated cucumber ‘Moskovsky Teplichnyi’ (Cucumis sativus) seedlings increased as DLI increased from ~3 to ~13 mol·m‒2·d‒1 (Chermnykh and Kosobrukhov, 1987). Models to predict plant growth and development in response to MDT and DLI have been generated for many floriculture crops including celosia ‘Gloria Mix’ (Celosia argentea; Pramuk and Runkle, 2005), cyclamen ‘Metis Scarlet Red’ (Cyclamen persicum; Oh et al., 2015), impatiens ‘Accent Red’ (Impatiens walleriana; Pramuk and Runkle, 2005), marigold ‘Bonanza Yellow’ (Tagetes patula; Moccaldi and Runkle, 2007), pansy ‘Universal Violet’ (Viola ×wittrockiana; Adams et al., 1997a, b), petunia ‘Easy Wave Coral Reef’ and ‘Wave Purple’ (Petunia ×hybrida; Blanchard et al., 2011), and salvia ‘Vista Red’ (Salvia splendens; Moccaldi and Runkle, 2007). While models including single environmental parameters have been generated to predict growth and development of culinary herbs, to our knowledge, multiparameter models including MDT and DLI have not been published; these data are integral in increasing production efficiencies, yield, and quality in CEs. Our objectives were to 1) determine the extent DLI and MDT influence the growth and development of purple basil, sage, spearmint (Mentha spicata), and sweet basil; and 2) determine 96 to what extent DLI and MDT influence purple basil color. We hypothesized the MDTopt of each species would increase as DLI increases. We also hypothesized there would be an interactive effect of DLI and MDT on purple basil color, where increasing MDT will result in a less purple, more green plant and increasing the DLI will increase the purple color more when the MDT is lower but have no effect at higher MDTs. Materials and Methods Plant production. Purple basil ‘Dark Opal’, common sage ‘Extrakta’, and sweet basil ‘Nufar’ seeds were sown in phenolic foam cubes (2.5 × 2.5 × 4 cm, Horticube XL; Smithers Oasis, Kent, OH) and the flats were placed in a greenhouse. Spearmint ‘Spanish’ vegetative stem-tip cuttings were harvested from stock plants and inserted immediately in phenolic foam cubes and placed in a glass-glazed greenhouse with a vapor pressure deficit of 0.3 kPa maintained with steam injection on a propagation bench for one week. Overhead mist was provided for 5 s when the integrated light intensity achieved 0.20 mol·m2·h-1 or after 60 min, whichever first occurred. The mist contained reverse-osmosis water supplemented with water-soluble fertilizer (MSU Plug Special 13N–2.2P–10.8K; Greencare Fertilizers, Inc.) and micronutrients (M.O.S.T.; JR Peters, Inc., Allentown, PA) providing (mg·L–1) 60 nitrogen (N), 10 phosphorous (P), 50 potassium (K), 28 calcium (Ca), 5 magnesium (Mg), 27 sulfur (S), 16 iron (Fe), 10 zinc (Zn), 17 manganese (Mn), 5 copper (Cu), 3 boron (B), and 0.2 molybdenum (Mo). Seeds and seedlings were irrigated overhead daily with reverse osmosis water supplemented with 12N-1.8P-13.4K water-soluble fertilizer as reported in Chapter 2. MDT (~23 °C) was measured by 0.13-mm type E thermocouples (Omega Engineering). High-pressure sodium (HPS) lamps provided a 97 photosynthetic photon flux density (PPFD) of ~80 µmol·m–2·s–1, as measured with a quantum sensor (LI-190R Quantum Sensor; LI-COR Biosciences, Lincoln, NE) every 15 s, and means were logged every hour by a CR-1000 datalogger (Campbell Scientific, Logan, UT) to create a 16-h photoperiod and maintain a target DLI of 10 mol·m‒2·d‒1. On 6 Sept. 2017 (rep 1, sweet basil), 20 Oct. 2017 (rep 2, sweet basil), 19 Apr. 2018 (rep 1, purple basil, sage, and spearmint), and 30 Oct. 2018 (rep 2, purple basil, sage, and spearmint), two (spearmint and sweet basil), three (purple basil) or four (sage) weeks after sowing or sticking, the 12 seedlings or rooted cuttings of each crop per treatment were transplanted into hydroponic systems (Active aqua premium high-rise flood table; Hydrofarm) in five connecting glass-glazed greenhouse compartments with target MDTs of constant 23, 26, 29, 32, or 35 °C. Each greenhouse contained three hydroponic systems under 0%, ~30%, or ~50% shade cloth (Solaro 3215 D O FB and Solaro 5220 D O; Ludvig Svensson, Kinna, Sweden) used to create target DLIs of 12, 9, or 7 mol·m‒2·d‒1, respectively. The production systems and cultural and environmental control and monitioring are the same as Chapter 2, unless reported. Growth data collection and analysis. The experiment was organized in a split-plot design with each of five MDTs in separate greenhouse sections and three DLI treatments in each section repeated in time. Plants were harvested when one treatment reached individual marketable size, which was three, four, or five weeks after transplant for spearmint and sweet basil, purple basil, or sage, respectively. The most recent fully expanded leaf of five purple basil, sage, and spearmint in each temperature and DLI treatment were dark acclimated for >15 minutes using manufacturer-supplied clips. Dark- acclimated leaves were exposed to 3,500 µmol·m–2·s–1 of red radiation (peak wavelength 650 98 nm) to saturate photosystem II, fluorescence was measured, and maximum quantum yield of dark-adapted leaves (Fv/Fm) was calculated and reported by a portable chlorophyll fluorescence meter (Handy Plant Efficiency Analyzer (PEA); Hanstech Instruments Ltd. Norfolk, UK). The International Commission on Illumination (CIE) L*a*b color space values of purple basil were measured using a colorimeter (Chroma Meter CR-400; Konica Minolta Sensing; Tokyo, Japan) to quantify foliage coloration. L* is a measure of lightness ranging from 0 (black) to 100 (white), and a* and b* are interpreted on a positive and negative scale. The a* scale ranges from red (positive values) to green (negative values), while b* ranges from yellow (positive values) to blue (negative values). However, these variables are not independent and therefore, cannot be interpreted independently, so Chroma (C*) and hue angle (h°) were calculated (McGuire, 1992). Chroma (C*) is the degree of departure from gray toward pure chromatic color, calculated as √"∗"+$∗" and representing the hypotenuse of an a*b* plot (McGuire, 1992). Hue angle (h°) is the angle from 0° or 360° on the color wheel. Values 0° and 360° indicate red, 90° is yellow, 120° is green, 180° is bluish green, and 270° is blue. To account for positive and negative a* and b* values, the following equations were used based on McLellan et al. (1994): "( )360 If a* and b* are positive: ℎ°=(#$#%&!∗#∗' "( )360 If a* is positive and b* is negative: ℎ°=360+(#$#%&!∗#∗' If a* is negative and b* is positive, or if a* and b* are negative: ℎ°=180+(#$#%&!∗#∗' "( )360 Since h° is a continuous circular scale and h° values fell between 0° and 127° or 300° and 360°, values over 300° were transformed by subtracting 360° for data analysis. 99 The number of branches >2.5 cm and height from the substrate surface to the tip of the tallest leaf were recorded, and leaf area of four most recent fully expanded leaves was measured with a with a leaf-area meter (LI-300; LI-COR Biosciences) for 10 plants per treatment. Leaves were separated from stems, and leaf and stem fresh mass of 10 plants per treatment were recorded. Leaf fresh mass fraction was calculated as leaf fresh mass/total fresh mass. Tissue was placed in a forced-air oven maintained at 75 ºC for at least 3 d, weighed, and dry mass was recorded. Dry matter concentration (DMC) was calculated as g dry mass per kg fresh mass. Analysis of variance was performed using JMP (version 12.0.1, SAS Institute Inc., Cary, NC) and when interactions were not significant, data were pooled. Linear, quadratic, and surface regression analyses were conducted using SigmaPlot (version 11.0, Systat Software Inc., San Jose, CA). Equations used to generate predictive models were based on 300 observations for each species. Results Purple basil. As MDT increased from 22.9 to 35.0 °C and DLI increased from 5.0 to 18.7 mol·m‒2·d‒1, purple basil branch number increased linearly by 18 and 12 branches, respectively (Table III-2, Figs. III-1A-B). DMC decreased linearly by 8% (5.6 g·kg‒1) as MDT increased from 22.9 to 35.0 °C and increased quadratically as DLI increased from 5.0 to 18.7 mol·m‒2·d‒1, with a near-linear increase from 5.0 to 14 mol·m‒2·d‒1 (61 to 74 g·kg‒1) and similar DMCs as DLI further increased to 18.7 mol·m‒2·d‒1 (76 g·kg‒1; Table III-2, Fig. III-1C-D). As MDT increased from 22.9 to 35.0 °C, fresh mass of purple basil increased quadratically 4.3-fold (Fig. III-1E). As DLI increased from 5.0 to 18.7 mol·m‒2·d‒1, fresh mass increased linearly over 3-fold (by 76 g; Fig. III-1F). 100 Height of purple basil at harvest increased as MDT increased, but the magnitude was dependent upon DLI and more pronounced at lower DLIs (Table III-2, Fig. III-2A). Plants were 11.4 and 5.6 cm taller as MDT increased from 22.9 to 35.0 °C when the DLI was 5.0 and 18.7 mol·m‒2·d‒1, respectively. The maximum quantum yield of dark-adapted leaves (Fv/Fm) was highest (0.80) when DLI was low (5.0 mol·m‒2·d‒1) and MDT was high (35.0 °C; Table III-2, Fig. III-2B). Reducing MDT or increasing DLI decreased Fv/Fm by up to 0.03 and 0.06, respectively. Fv/Fm values were similar (0.74 to 0.75) regardless of MDT when DLI was high (18.7 mol·m‒2·d‒1). Smaller leaves developed as MDT increased (Table III-2, Fig III-2C). For instance, increasing MDT from 22.9 to 35.0 °C reduced leaf area (of the four leaves measured) by 81 cm2 (60%) when DLI was 18.7 mol·m‒2·d‒1. Increasing DLI increased leaf area more when the MDT was lower (22.9 °C; 46 cm2 increase) than when the MDT was higher (35.0 °C; 10 cm2 increase). MDT and DLI interacted to influence the fraction of total fresh mass comprised of leaves compared to stems (Table III-2, Fig. III-2D). The greatest proportion of leaves occurred when both DLI and MDT were low. When DLI was 5.0 mol·m‒2·d‒1 and MDT was 22.9 °C, the leaf mass fraction was 88%; increasing MDT to 35.0 °C decreased leaf mass fraction to 77% to 78% regardless of DLI. The interaction of MDT and DLI influenced h° (Table III-2, Figs. III-3A and III-4). Purple basil was the reddest/bluest when grown with a high DLI (18.7 mol·m‒2·d‒1) and low MDT (22.9 °C). When MDT was 24.3 °C and DLI was 13.7 mol·m‒2·d‒1, h° was 333° (2° calculated). As MDT increased to 35.0 °C, color became greener (138°). The influence of DLI on color was MDT-dependent. Decreasing DLI from 18.7 to 5.0 mol·m‒2·d‒1 changed h° from 335° (-25°) to 98° when MDT was 25 °C, but remained green (99° to 138°) regardless of DLI when MDT was 35.0 °C. Purple basil plants were lighter (higher L*) as DLI decreased or MDT 101 increased (Table III-2, Fig. III-3B). L* was influenced by the interaction of MDT and DLI; L* increased more in response to decreased DLI at lower MDTs. For example, as DLI decreased from 18.7 to 5.0 mol·m‒2·d‒1, L* increased from 28 to 33 at a MDT of 22.9 °C and from 39 to 41 at a MDT of 35.0 °C. Increasing MDT increased the distance away from greytone linearly; C* increased from 0.7 to 17.3 as MDT increased from 22.9 to 35.0 °C (Table III-2). Sage. MDT and DLI interacted to influence sage branch number, height, leaf fresh mass fraction, and fresh mass (Table III-2, Fig. III-5A‒D). Branch number increased by 7 (70%) as MDT decreased from 35.1 to 22.8 °C when DLI was 18.3 mol·m‒2·d‒1, but branch number only increased by 2 (23%) when DLI was 4.9 mol·m‒2·d‒1 (Table III-2, Fig. III-5A). Under a DLI of 18.3 mol·m‒2·d‒1, plants were 4-cm taller at a MDT of 30.0 °C compared to 22.8 °C, but were 14-cm shorter when grown at 35.1 °C than 30.0 °C. The height decrease was attenuated when DLI was lower (4.9 mol·m‒2·d‒1) when plants were 11-cm shorter as MDT increased from 30.0 °C to 35.1 °C. Unlike purple basil, the greatest proportion of fresh mass residing in sage leaves occurred when both DLI and MDT were high. At a DLI of 18.3 mol·m‒2·d‒1 and MDT of 35.1 °C, the leaf mass fraction was 79% and decreased to 69% at a DLI of 8.1 and MDT of 24.4 °C. Fresh mass increased as DLI increased or MDT decreased. At a MDT of 22.8 °C, increasing DLI from 4.9 to 18.3 mol·m‒2·d‒1 increased fresh mass 2.4-fold (by 40.6 g); at a MDT of 35.1 °C, the same increase in DLI increased fresh mass 90% (by 8.8 g). Similarly, when DLI was 18.3 mol·m‒2·d‒1, decreasing MDT from 35.1 to 22.8 °C increased fresh mass 2.1-fold (by 39.1 g) while the same decrease in MDT when DLI was 4.9 mol·m‒2·d‒1 increased fresh mass by 75% (7.3 g). 102 DMC increased linearly as both DLI and MDT increased, and was not influenced by their interaction (Table III-2, Figs. III-6A-B). As MDT increased from 22.8 to 35.1 °C and DLI increased from 4.9 to 18.3 mol·m‒2·d‒1, DMC increased by 30 g·kg‒1 and 56 g·kg‒1, respectively. Leaf area (of the four leaves measured) decreased linearly by 102 cm2 (67%) as MDT increased from 22.8 to 35.1 °C (Table III-2, Fig. III-6C). Similarly, the same increase in MDT decreased Fv/Fm linearly from 0.84 to 0.80 (4%; Table III-2, Fig. III-6D) while DLI did not influence leaf area or Fv/Fm. Spearmint. Spearmint branch number, height, Fv/Fm, leaf area (of the four leaves measured), dry matter concentration, leaf fresh mass fraction, and fresh mass were influenced by the interaction of MDT and DLI (Table III-2, Figs. III-7A‒G). Plants were generally taller, had more branches, and larger leaves as MDT increased, but the magnitude of the MDT effect depended upon DLI (Table III-2, Figs. III-7A, B, and D). When DLI was 6.1 mol·m‒2·d‒1, increasing MDT from 23.0 to 35.1 °C increased branch number by 4 (40%) and height by 15.7 cm (65%) and decreased leaf area by 38 cm2 (42%). When DLI was 19.4 mol·m‒2·d‒1, the same MDT increase increased branch number by 2 (17%) and decreased height by 2.2 cm (7%) and leaf area by 88 cm2 (67%). Fv/Fm was greatest when MDT was high (35.1 °C) and DLI was low (6.1 mol·m‒2·d‒1; Table III- 2, Fig. III-7C). Increasing MDT from 23.0 °C to 35.1 °C increased DMC by 16 g·kg‒1 (15%) when DLI was 19.4 mol·m‒2·d‒1 but hardly influenced DMC when DLI was 6.1 mol·m‒2·d‒1 (Table III-2, Fig. III-7E). When DLI was 6.1 mol·m‒2·d‒1, leaf mass fraction decreased from 67% to 62% as MDT increased from 23.0 to 35.1 °C, but when DLI was 19.4 mol·m‒2·d‒1, the same increase in MDT decreased it from 59% to 57% (Table III-2, Fig. III-7F). Fresh mass 103 increased by 60 g as DLI increased from 6.1 to 19.4 mol·m‒2·d‒1 when MDT was 23.0 °C and increased by 32 g when MDT was 35.1 °C (Table III-2, Fig. III-7G). Sweet basil. Sweet basil branch number, fresh mass, height, leaf area (of the four leaves measured), and node number responded somewhat similarly to MDT and DLI (Table III-2, Fig. III-8A-E). Plants were taller, weighed more, had more branches, and had larger leaves as MDT increased, but the magnitude of the MDT effect depended upon DLI. When DLI was 5.5 mol·m‒2·d‒1, increasing MDT from 23.0 °C to 35.7 °C increased branch number by 7 (from 0), fresh mass by 15 g (3-fold), height by 6.5 cm (48%), and nodes by 2 (39%), and decreased leaf area by 22 cm2 (8%). When the DLI was 13.2 mol·m‒2·d‒1, the same MDT increase increased branch number by 14 (6.2-fold), fresh mass by 37 g (123%), height by 10.4 cm (65%), and nodes by 3 (62%), and decreased leaf area by 248 cm2 (53%). Fv/Fm was lowest at 24.4 °C (0.77) and was similar (0.81 to 0.83) at higher temperatures (25.6 to 35.7 °C; data not shown) while DLI did not influence Fv/Fm. Sweet basil DMC increased by 6 g·kg‒1 (8%) as DLI increased from 5.5 to 13.2 mol·m‒ 2·d‒1 and was not influenced by MDT (Table III-2). Discussion Plant development. Plant development, including leaf unfolding rate and progression towards flowering, are primarily determined by temperature since it influences the rate of meristematic differentiation. Development can also be influenced by a radiation-induced temperature response (increased meristem temperature) and DLI (increased carbon substrate availability). In the current study, increasing MDT increased branch number of all species grown (Figs. III-1A, III-5A, III-7A, and 104 III-8A). Similarly, Chang et al. (2005) found that basil grown at 25 or 30 °C had two more branches than those grown at 15 °C. Sweet basil ‘Improved Genovese Compact’ branches formed more quickly when the DLI was higher (Dou et al., 2018). In our study, sweet basil branch number also increased as DLI increased, but the effect was attenuated when MDT was low (23 °C; Fig. III-8A). In contrast, sage and spearmint branch number also increased with increasing DLI, but more so when MDT was lower (23 °C) compared to 35 °C; purple basil branch number was not influenced by DLI (Fig. III-5A and III-7A). While some developmental responses may be initiated or regulated by radiation, increasing radiation also increases shoot temperature (Faust and Heins, 1998). This results in a radiation-induced temperature response, especially when meristem temperature is impacted, influencing cell differentiation and development. The height of plants with a central stem is a function of growth and development; both leaf unfolding rate (node number) and internode elongation contribute to height. Many researchers have documented that as DLI and temperature increased, basil height increased (Chang et al., 2005; Dou et al., 2018; Mortensen, 2014; Putievsky, 1983). For instance, sweet basil ‘Improved Genovese Compact’ height increased from 17.4 to 23.3 cm as DLI increased from 9.3 to 16.5 mol·m‒2·d‒1 when grown for 21 d after transplant under sole-source lighting (Dou et al., 2018). Mortensen (2014) reported that as temperature increased from 18 to 27 °C, the height of basil increased from 8 to 18 cm. In a separate study, Chang et al. (2005) found that basil grown at 15 °C were shorter than those grown at 25 or 30 °C but there was no difference in height between basil grown at 25 or 30 °C. Lastly, as daytime temperature increased from 18 to 30 °C, basil was taller at 90 and 120 days post-germination, but differences were not apparent by 105 150 d (Putievsky, 1983). For sage, plant height increased from 9 to 18 cm as temperature increased from 18 to 27 °C (Mortensen, 2014). MDT and DLI have been reported to interact and influence height of ornamentals. In campanula ‘Blue Clips’ (Campanula carpatica), height at flowering was not influenced by MDT, but by the difference in day and night temperature (Niu et al., 2001). The magnitude of difference in plant height to the changing temperature was stronger when the DLI was lower than under medium to high DLIs (Niu et al., 2001). In the current study, as MDT increased, purple basil, sage, sweet basil, and spearmint height increased to a DLI-dependent maximum height (Figs. III-2A, III-5B, III-7B, and III-8C). For example, purple basil height increased by 11.4 cm as MDT increased from 22.9 to 35 °C when DLI was 5 mol·m‒2·d‒1 but only 5.6 cm when DLI was 18.7 mol·m‒2·d‒1 (Fig. III-2A). Low Fv/Fm is an indicator of plant stress resulting from an inefficiency of photosystem II. Purple basil, spearmint, and sage tended to exhibit lower Fv/Fm values as MDT decreased. Sweet basil had the lowest Fv/Fm at 24.4 °C, but similarly higher Fv/Fm among higher MDTs (Figs. III- 2B, III-6D, and III-7C). Similar trends have been demonstrated in sweet basil ‘Nufar’, lemon basil ‘Sweet Dani’ and ‘Lime’, and holy basil, where MDTs from ~11 to ~23 °C had reduced Fv/Fm compared to higher MDTs up to 35 °C (Walters and Currey, 2019). For purple basil and spearmint, the MDT induced reduction in Fv/Fm was more pronounced when the DLI was lower (Figs. III-2B and III-7C). Growth, mass concentration, and partitioning. Our observation of decreased leaf area (of the four leaves measured) with increased MDT across all four crops (Figs. III-2C, III-6D, III-7D, and III-8D) has been documented in many species and cultivars including spinach ‘Savoy’ (Spinacia oleracea; Boese and Huner, 1990), 106 five sunflower (Helianthus annuus) cultivars (Rawson and Hindmarsh, 1982), and several tomato (Solanum lycopersicum) cultivars (de Koning, 1994). Additionally, DLI can influence leaf area. Dou et al. (2018) reported leaf area of sweet basil ‘Improved Genovese Compact’ plants increased from 406 to 614 cm2 as DLI increased from 9.3 to 17.8 mol·m‒2·d‒1. However, the specific leaf area, or area per unit leaf dry weight, decreased from 518 to 398 cm2·g‒1 as DLI increased. The response of purple basil, spearmint, and sweet basil leaf area to MDT was stronger at higher DLIs, and leaf area tended to increase quadratically as DLI increased (Figs. III-2C, III-7D, and III-8D). Leaf area of geranium ‘Sooner Red’ (Pelargonium × hortorum) increased as DLI increased from <5 to 20 mol·m‒2·d‒1 (Armitage et al., 1981) but decreased in geranium ‘Red Elite’ and heliconia ‘Golden Torch’ (Heliconia psittacorum × spathocircinata) as DLI increased from ~20 to ~30 mol·m‒2·d‒1 (Catley and Brooking; 1996; Korczynski et al., 2002; White and Warrington, 1984). Dry matter concentration is an indication of plant quality; higher DMC is desired and is an indication of the commercially used term “toning” (Faust et al., 2005). In general, DMC increased as DLI increased in our study, but the influence of MDT on DMC was crop dependent. As MDT increased, purple basil DMC decreased while sage and spearmint DMC increased (Figs. III-1C, III-6A, III-7E). Similar to sage and spearmint, Dou et al. (2018) reported DMC of sweet basil ‘Improved Genovese Compact’ increased linearly from 67 to 92 g·kg‒1 as DLI increased from 9.3 to 17.8 mol·m‒2·d‒1. In another study, water content of mint, sage, and sweet basil decreased by 5.2%, 6.1%, and 2.7% as DLI increased from 2 to 20 mol·m‒2·d‒1 (Litvin, 2019), and DMC of ornamentals including ageratum ‘Hawaii White’ (Ageratum houstonianum), begonia ‘Vodka Cocktail’ (Begonia ×semperflorens-cultorum), impatiens ‘Cajun Red’, marigold ‘American Antigua Orange’, petunia ‘Apple Blossom’, salvia ‘Lady in Red’ (Salvia coccinea), 107 vinca ‘Pacific Lilac’ (Catharanthus roseus), and zinnia ‘Dreamland Rose’ (Zinnia elegans) increased linearly or quadratically as DLI increased from 4.8 to 42.9 mol·m‒2·d‒1 (Faust et al., 2005). In a meta-analysis of biomass allocation patterns to leaves, stems, and roots, Poorter et al. (2011) determined that the fraction of whole-plant mass represented by leaves most strongly decreased as DLI increased up to ~20 mol·m‒2·d‒1, above which the response saturated. Additionally, increasing temperature increased leaf mass fraction, though the response was attenuated compared to increasing DLI (Poorter et al., 2011). In our study, MDT and DLI interacted to influence leaf mass fraction. Congruent with the meta-analysis, increasing sage and spearmint DLI and sage MDT increased leaf mass fraction quadratically (Figs. III-5C and III- 7F). In contrast, increasing purple basil DLI and spearmint and purple basil MDT decreased leaf mass fraction quadratically (Figs. III-2D and III-7F). Fresh mass modeling. A temperature response curve can be utilized to calculate fresh mass in response to MDT. For example, as MDT increased from 18 to 35 °C, dry mass MDTopt of six sage species (Salvia spp.) occurred at MDTs ranging from 21 to 35 °C but varied among species (Lasseigne et al., 2007). For common sage, fresh mass increased from 37 to 72 g as temperature increased from 18 to 27 °C (Mortensen, 2014). The influence of MDT on sweet basil fresh mass has been studied extensively (Chang et al., 2005; Pogany et al., 1968; Putievsky, 1983; Walters and Currey, 2019). The Tb has been reported to range from 10.9 to 12.1 °C, and the Topt has been found to range from 25 to 35 °C (Chang et al., 2005; Pogany et al., 1968; Putievsky, 1983; Walters and Currey, 2019). Purple basil fresh mass in our study increased quadratically as MDT increased from 22.9 to 35.0 °C (Fig. III-1E). Although supraoptimal MDTs were not observed and the 108 calculated MDTopt is beyond the experimental range, using the regression equation, the extrapolated MDTopt is 37.2 °C; however, this prediction may not be accurate. While similar trends in sweet basil were observed, in that fresh mass increased as MDT increased from 23 to ~35 °C, we determined that MDT and DLI interacted not only to influence sweet basil, but also sage and spearmint fresh mass (Figs. III-5D, III-7G, and III-8E). In general, fresh mass of purple basil, sage, spearmint, and sweet basil increased either linearly or quadratically as DLI increased (Figs. III-1F, III-5D, III-7G, and III-8E). Purple basil fresh mass increased linearly from 21.8 to 97.4 g as DLI increased from 5.0 to 18.7 mol·m‒2·d‒1. Other researchers observed similar trends; as DLI increased from 9.3 to 17.8 mol·m‒2·d‒1, sweet basil ‘Improved Genovese Compact’ fresh mass increased linearly from 13.1 to 23.3 g (Dou et al., 2018) and increasing DLI from 2 to 20 mol·m‒2·d‒1 increased sweet basil ‘Nufar’ fresh mass linearly by 136.4 g (Litvin, 2019). Mint and sage fresh mass increased as DLI increased from 2 mol·m‒2·d‒1 to a DLIopt of 14.9 and 15.9 mol·m‒2·d‒1, respectively (Litvin, 2019). Since MDT and DLI interacted to influence sage, spearmint, and sweet basil fresh mass, determining MDTopt or DLIopt based on other environmental factors can improve production efficiency to achieve desired growth and energy-use efficiency outcomes, especially since the optimal MDT or DLI may shift as environmental factors interact. Additionally, the ability to calculate MDTopt and DLIopt requires supraoptimal temperatures. Since supraoptimal temperatures were achieved for many parameters measured in this study, fresh mass was chosen for this example, but MDTopt and DLIopt could also be calculated for other parameters. Therefore, similar to the methods used in Chapter 2, surface regression models (Table III-2, Figs. III-5D, III-7G, and III-8E) were used to calculate the MDTopt for each parameter based on DLI by using /M!"opt = " + 2c(MDT) + 0(DLI) = 0, and the DLIopt based on MDT using /!#$opt = $ + 21(DLI) + 109 0(MDT) = 0 (Figs. III-9A-E). For example, as DLI increased from 6 to 18 mol·m‒2·d‒1, sage and spearmint fresh mass MDTopt decreased from 27.5 to 23.5 °C and from 30.4 to 27.8 °C, respectively (Figs III-9A and E). In contrast, as DLI increased from 6 to 11 mol·m‒2·d‒1, sweet basil fresh mass MDTopt increased from 32.6 to 35.5 °C (Fig. III-9C). Similarly, as MDT increased from 24 to 32 °C, sage fresh mass DLIopt decreased from 17.9 to 14.4 mol·m‒2·d‒1 while sweet basil fresh mass DLIopt increased from 11.8 to 13.2 mol·m‒2·d‒1 (Figs. III-9B and D). The DLIopt for spearmint and MDTopt for purple basil could not be calculated because supraoptimal DLIs and MDTs, respectively, were not achieved. Color. For purple basil, color is an additional quality parameter to consider in addition to plant mass that contributes not only to visual appeal but is also an indicator of anthocyanin concentration (Phippen and Simon, 1988). Anthocyanins are desired in food crops due to their antioxidant activity, potentially increasing the nutritional value when consumed, and their contribution to the red and purple color of crops. Anthocyanins function largely as defense molecules. For example, increased anthocyanin concentration can impart freezing tolerance or protect against excess radiant energy (Boldt, 2014.) As described previously, temperature plays a large role in the kinetics of plant enzymatic reaction rates. In general, as temperature increases above Tb, enzymatic activity increases. Beyond an enzymatic-dependent Tmax, enzymes begin to denature and cease in their functions. However, enzymes can also be up- or down-regulated by either increases or decreases in temperature. One major enzyme in the anthocyanin biosynthesis pathway, phenylalanine ammonia lyase (PAL), is upregulated by decreasing temperatures (Leyva et al., 1995). As temperatures decrease, anthocyanin biosynthesis and accumulation increases (Kim et al., 2017). 110 This is mainly due to reduced anthocyanin degradation (Rehman et al., 2017) and reduced high- temperature dependent degradation of ELONGATED HYPOCOTYL 5 (HY5) protein, a positive regulator of anthocyanin biosynthesis at multiple points in the pathway (Kim et al., 2017). In crops such as grape (Vitis vinifera) and apple (Malus domestica), researchers have reported reduced anthocyanin concentrations as temperatures increase (Lin-Wang et al., 2011; Mori et al., 2007; Movahed et al., 2016; Rehman et al., 2017). In the current study, increasing purple basil MDT (regardless of DLI) increased the green coloration quantified by a ~120 h°, low a*, and high b* (Table III-2, Figs. III-3A and III-4). The greener plants were also lighter (L*) colored (Fig. III-3B). Anthocyanins can also accumulate in response to increased DLI or radiation intensities, functioning as photoprotectants (Boldt, 2014). When MDT is low, increasing DLI is an effective strategy to increase the red/blue (purple) color quantified by a ~260 h°, increased a*, and reduced b* (Fig. 3A). In addition to increasing the greenness of purple basil, increasing MDT also increased the C* or magnitude of colorfulness (Table III-2). If temperature is the primary environmental tool a grower can manipulate to enhance fresh mass yield, product quality should be taken into account. If yield is the main production factor and purple coloration is secondary, green sweet basil cultivars could be grown instead of purple basil cultivars due to their much higher growth rate (Walters and Currey, 2015). Conclusions These data and models generated allow us to predict plant growth, development, and leaf coloration in response to MDT and DLI and to conduct energy-benefit analyses. Though technology to manipulate the growing environment exists, these data will allow growers to 111 calculate the most advantageous growing environment (taking growth, development, quality, and energy into account) and utilize their technologies to realize it. In addition to aiding growers in manipulating the environment for specific culinary herb crops, comparisons can be made to group or separate plants based on growth rates and favorable production environments. Different cultivars may respond differently to MDT and DLI. These models can serve as a basis for environmental controls with general trends informing production decisions and estimating increased (or decreased) growth, development, and quality potential. Additionally, these data can contribute to future multi-parameter machine-learning models. 112 APPENDIX 113 Table III-1. Average daily light integral (DLI; mol·m‒2·d‒1 ± SD) and mean daily air temperature (MDT), leaf, and nutrient solution temperatures (°C ± SD) over the three (spearmint and sweet basil), four (purple basil), or five (sage) week growing period for two replications over time. Plants were transplanted on 6 Sept. 2017 (rep 1, sweet basil), 20 Oct. 2017 (rep 2, sweet basil), 19 Apr. 2018 (rep 1, purple basil, sage, and spearmint), and 30 Oct. 2018 (rep 2, purple basil, sage, and spearmint). Data were collected every 15 s with means logged every hour. DLI Air Temperature °C Leaf Solution Spearmint ‘Spanish’ (Mentha spicata) - - - z - - 22.5 ± 1.3 14.3 ± 2.5 24.3 ± 1.4 22.3 ± 1.8 13.7 ± 2.6 24.3 ± 1.4 9.7 ± 1.7 24.3 ± 1.4 22.3 ± 1.5 19.1 ± 3.5 26.6 ± 0.9 27.2 ± 1.5 24.4 ± 0.9 24.2 ± 0.9 13.0 ± 2.5 26.6 ± 0.9 8.7 ± 1.8 26.6 ± 0.9 24.1 ± 1.0 18.6 ± 3.5 28.9 ± 1.1 28.7 ± 0.9 25.7 ± 0.7 25.2 ± 0.4 13.4 ± 2.5 28.9 ± 1.1 25.1 ± 0.5 9.3 ± 1.7 28.9 ± 1.1 29.0 ± 1.4 19.4 ± 3.8 32.5 ± 0.4 13.6 ± 2.7 32.5 ± 0.4 27.3 ± 1.0 9.7 ± 1.7 32.5 ± 0.4 26.3 ± 1.2 16.9 ± 3.0 35.1 ± 0.7 35.8 ± 0.8 30.6 ± 0.6 12.2 ± 2.1 35.1 ± 0.7 30.0 ± 0.6 9.9 ± 1.6 35.1 ± 0.7 29.5 ± 0.7 10.4 ± 0.9 23.0 ± 1.2 23.1 ± 1.6 22.2 ± 0.9 21.1 ± 0.9 8.7 ± 0.9 23.0 ± 1.2 6.5 ± 0.8 23.0 ± 1.2 20.8 ± 0.8 10.4 ± 1.3 26.1 ± 0.4 27.0 ± 0.4 23.3 ± 0.6 7.4 ± 0.8 26.1 ± 0.4 22.8 ± 0.5 - - - - - - - - - - Temperature °C - - Air - - - Leaf DLI Solution Purple basil ‘Dark Opal’ (Ocimum basilicum) 22.7 ± 1.4 13.7 ± 2.6 24.3 ± 1.6 22.4 ± 1.8 13.0 ± 2.6 24.3 ± 1.6 9.3 ± 1.7 24.3 ± 1.6 22.4 ± 1.5 18.6 ± 3.5 26.5 ± 0.9 27.0 ± 1.8 24.4 ± 0.9 24.1 ± 1.0 12.5 ± 2.6 26.5 ± 0.9 8.2 ± 1.9 26.5 ± 0.9 23.9 ± 1.2 18.0 ± 3.4 28.6 ± 1.1 28.5 ± 1.0 25.8 ± 0.7 25.1 ± 0.4 12.9 ± 2.6 28.6 ± 1.1 25.1 ± 0.5 9.0 ± 1.7 28.6 ± 1.1 29.4 ± 1.5 18.7 ± 3.8 32.4 ± 0.4 13.0 ± 2.7 32.4 ± 0.4 27.3 ± 0.9 9.3 ± 1.7 32.4 ± 0.4 26.4 ± 1.2 16.2 ± 3.1 35.0 ± 0.7 35.9 ± 0.7 30.6 ± 0.6 11.8 ± 2.1 35.0 ± 0.7 30.1 ± 0.6 9.6 ± 1.5 35.0 ± 0.7 29.5 ± 0.7 10.3 ± 1.0 22.9 ± 1.1 22.6 ± 1.6 22.0 ± 0.9 20.8 ± 1.0 8.6 ± 0.9 22.9 ± 1.1 6.4 ± 0.8 22.9 ± 1.1 20.6 ± 0.8 10.7 ± 1.3 26.1 ± 0.4 27.1 ± 0.5 23.3 ± 0.5 7.3 ± 0.8 26.1 ± 0.4 22.8 ± 0.4 - - - - - - - - - - 114 Rep. 1 2 Table III-1 (cont’d). 1 2 - - - 22.6 ± 0.3 5.0 ± 0.8 26.1 ± 0.4 9.7 ± 1.0 28.7 ± 0.6 29.8 ± 2.3 25.2 ± 0.4 24.9 ± 0.5 7.9 ± 0.8 28.7 ± 0.6 6.8 ± 0.9 28.7 ± 0.6 24.2 ± 0.5 11.9 ± 1.2 32.1 ± 1.5 30.1 ± 0.5 27.7 ± 0.3 27.7 ± 1.1 8.8 ± 0.8 32.1 ± 1.5 6.1 ± 0.8 32.1 ± 1.5 27.7 ± 1.1 11.2 ± 1.0 34.5 ± 0.8 35.4 ± 1.7 30.0 ± 0.8 8.5 ± 0.8 34.5 ± 0.8 29.4 ± 0.5 29.1 ± 1.6 y 6.6 ± 0.8 34.5 ± 0.8 - - - - Sage ‘Extrakta’ (Salvia officinalis) - - - - - 13.3 ± 2.6 24.5 ± 1.5 22.9 ± 1.4 12.6 ± 2.8 24.5 ± 1.5 22.8 ± 1.8 9.1 ± 1.6 24.5 ± 1.5 22.7 ± 1.5 18.3 ± 3.5 26.6 ± 0.9 27.1 ± 1.8 24.6 ± 1.0 12.2 ± 2.7 26.6 ± 0.9 24.1 ± 1.0 8.0 ± 2.0 26.6 ± 0.9 24.0 ± 1.4 17.6 ± 3.5 28.6 ± 1.1 28.5 ± 1.1 26.0 ± 0.7 12.5 ± 2.6 28.6 ± 1.1 25.2 ± 0.4 8.7 ± 1.7 28.6 ± 1.1 25.2 ± 0.5 18.3 ± 3.8 32.5 ± 0.4 29.8 ± 1.6 12.7 ± 2.7 32.5 ± 0.4 27.5 ± 0.9 9.1 ± 1.7 32.5 ± 0.4 26.6 ± 1.3 15.8 ± 3.3 35.1 ± 0.6 36.0 ± 0.8 30.8 ± 0.6 11.6 ± 2.2 35.1 ± 0.6 30.3 ± 0.7 9.5 ± 1.5 35.1 ± 0.6 29.7 ± 0.7 10.2 ± 1.1 22.8 ± 1.0 22.5 ± 1.5 22.0 ± 0.8 20.8 ± 0.9 8.5 ± 1.0 22.8 ± 1.0 6.4 ± 0.8 22.8 ± 1.0 20.5 ± 0.8 27.0 ± 0.5 23.3 ± 0.5 10.7 ± 1.3 26.1 ± 0.4 7.2 ± 0.8 26.1 ± 0.4 22.8 ± 0.4 - - - - - - - - - - - - - 22.5 ± 0.3 5.0 ± 0.7 26.1 ± 0.4 9.6 ± 1.0 28.7 ± 0.5 29.6 ± 2.0 25.2 ± 0.4 24.8 ± 0.5 7.9 ± 0.8 28.7 ± 0.5 6.8 ± 0.9 28.7 ± 0.5 24.1 ± 0.5 11.7 ± 1.4 32.1 ± 1.4 30.0 ± 0.5 27.9 ± 0.5 27.5 ± 1.1 8.7 ± 0.8 32.1 ± 1.4 6.1 ± 0.7 32.1 ± 1.4 27.5 ± 1.1 11.2 ± 1.0 34.4 ± 0.8 34.7 ± 2.0 29.8 ± 0.8 8.4 ± 0.8 34.4 ± 0.8 29.5 ± 0.6 28.7 ± 1.6 y 6.5 ± 0.8 34.4 ± 0.8 - - - - Sweet basil ‘Nufar’ (Ocimum basilicum) - - - - - - - 24.7 ± 0.9 y 12.5 ± 0.8 24.4 ± 1.9 24.3 ± 1.3 9.7 ± 0.9 24.4 ± 1.9 8.1 ± 0.8 24.4 ± 1.9 23.6 ± 1.7 13.1 ± 1.7 27.0 ± 0.8 30.1 ± 1.5 26.3 ± 0.8 25.7 ± 0.7 10.1 ± 1.0 27.0 ± 0.8 7.7 ± 0.7 27.0 ± 0.8 25.7 ± 0.7 13.0 ± 1.7 30.2 ± 1.2 30.7 ± 0.7 28.9 ± 0.9 28.4 ± 1.1 9.5 ± 1.3 30.2 ± 1.2 7.2 ± 1.0 30.2 ± 1.2 28.0 ± 1.1 11.7 ± 1.6 32.9 ± 0.8 37.0 ± 0.7 30.8 ± 0.4 9.3 ± 1.5 32.9 ± 0.8 30.3 ± 0.5 7.0 ± 0.9 32.9 ± 0.8 29.3 ± 0.6 12.7 ± 1.1 35.3 ± 1.0 37.4 ± 1.1 32.3 ± 1.0 9.4 ± 0.8 35.3 ± 1.0 31.6 ± 0.7 6.3 ± 0.5 35.3 ± 1.0 31.9 ± 1.0 12.9 ± 1.1 23.0 ± 1.3 26.3 ± 1.3 21.7 ± 1.4 21.5 ± 1.1 10.1 ± 1.0 23.0 ± 1.3 7.4 ± 0.8 23.0 ± 1.3 21.4 ± 1.3 11.1 ± 1.1 25.6 ± 0.6 27.7 ± 0.9 23.9 ± 0.7 9.2 ± 1.0 25.6 ± 0.6 23.4 ± 0.7 - z - - - - - - 115 4.9 ± 0.7 26.1 ± 0.4 9.5 ± 1.0 28.7 ± 0.5 7.8 ± 0.8 28.7 ± 0.5 6.7 ± 0.9 28.7 ± 0.5 11.5 ± 1.4 32.0 ± 1.3 8.5 ± 0.9 32.0 ± 1.3 6.0 ± 0.7 32.0 ± 1.3 11.0 ± 1.2 34.5 ± 0.7 8.3 ± 0.8 34.5 ± 0.7 6.5 ± 0.8 34.5 ± 0.7 Table III-1 (cont’d). z data not collected y partial data reported - - - 22.6 ± 0.3 29.4 ± 1.9 25.3 ± 0.4 24.9 ± 0.5 24.3 ± 0.5 29.9 ± 0.5 28.1 ± 0.6 27.5 ± 1.0 27.4 ± 1.0 34.6 ± 1.8 29.7 ± 0.8 29.7 ± 0.7 28.6 ± 1.4 y - - - - - 23.5 ± 0.7 8.0 ± 0.8 25.6 ± 0.6 11.3 ± 1.2 28.9 ± 1.9 30.5 ± 0.8 26.0 ± 1.0 25.9 ± 0.9 8.9 ± 1.1 28.9 ± 1.9 6.6 ± 0.9 28.9 ± 1.9 25.7 ± 1.0 13.2 ± 1.3 31.4 ± 0.7 35.7 ± 1.2 28.7 ± 0.8 27.3 ± 0.8 9.2 ± 1.0 31.4 ± 0.7 7.2 ± 0.7 31.4 ± 0.7 26.2 ± 0.8 12.0 ± 1.3 35.7 ± 0.8 36.6 ± 0.8 30.3 ± 0.8 8.5 ± 0.9 35.7 ± 0.8 29.6 ± 0.7 29.2 ± 0.6 5.5 ± 0.8 35.7 ± 0.8 - - - - - - 116 Table III-2. Regression analysis parameters and R2 or r2 for purple basil, sage, spearmint, and sweet basil branch number, height, maximum quantum yield of dark-adapted leaves (Fv/Fm), leaf area of four most recently fully expanded leaves, dry matter concentration (DMC), leaf mass fraction, and fresh mass in response to mean daily temperature (MDT; °C) and daily light integral (DLI; mol·m‒2·d‒1). All models are in the form of: ! = y0 + a*MDT + b*DLI + c*MDT2 (a) MDT Purple basil ‘Dark Opal’ (Ocimum basilicum) (c) MDT2 (b) DLI (d) DLI2 + d*DLI2 + e*MDT*DLI. Parameter y0 Branch (no.) Branch (no.) Height (cm) Fv/Fm Leaf area (cm2) DMC (g·kg‒1) DMC (g·kg‒1) Leaf mass fraction (fresh) Fresh mass (g) Fresh mass (g) Hue angle (h°) C* L* a* b* Branch (no.) Height (cm) -2.80E1 z (8.00) x 5.58 (3.72) -1.82E2 (1.47E1) 4.46E-1 (8.59E-2) -6.46E2 (7.61E1) 8.26E1 (8.61) 4.85E1 (4.92) 2.75 (1.30E-1) -3.58E2 (3.71E2) -5.83 (1.23E1) -9.94E2 (9.85E1) -3.09E1 (4.33) 5.33E1 (7.98) 2.32E1 (2.82) -3.07E1 (3.35) 1.92E1 (6.90) -2.02E2 (2.40E1) 1.46 (0.272) 1.24E1 (9.89E-1) 2.11E-2 (5.9E-3) 5.21E1 (5.22) -4.6E-1 (2.93E-1) -1.18E-1 (8.7E-3) 2.35E1 (2.58E1) -1.94E-1 (1.74E-2) -3.E-4 (1.E-4) -9.06E-1 (9.11E-2) 1.8E-3 (2.E-4) -3.15E-1 (4.42E-1) y 8.51E-1 (3.31E-1) 4.08 (5.89E-1) 6.1E-3 (3.2E-3) 1.11E1 (2.83) 2.99 (8.72E-1) -4.7E-2 (5.2E-3) 5.52 (1.09) -3.88E1 (3.67) -1.47 (1.18E-1) -6.99E-1 (2.97E-1) 8.23E1 (6.77) 1.379 (1.47E-1) -1.71 (5.47E-1) -9.19E-1 (9.60E-2) 1.32 (1.14E-1) 3.98E-2 (9.6E-3) Sage ‘Extrakta’ (Salvia officinalis) -1.31E-1 (2.59E-2) 1.45E-2 (8.0E-3) -2.67E-1 (2.84E-2) -8.45E-1 (4.73E-1) 1.55E1 (1.61) 1.103 (2.65E-1) 4.06 (9.87E-1) 117 -9.40E-2 (1.53E-2) -8.22E-2 (3.57E-2) 7.E-4 (1.E-4) (e) MDT*DLI R2 or r2 -3.52E-2 (1.81E-2) -3.E-4 (1.E-4) -3.38E-1 (9.56E-2) 9.E-4 (2.E-4) 1.19 (1.24E-1) 1.49E-2 (1.00E-2) -2.87E-2 (9.0E-3) -4.82E-2 (3.05E-2) 0.658 0.174 0.646 0.371 0.540 0.087 0.419 0.615 0.459 0.420 0.718 0.601 0.687 0.641 0.654 0.376 0.367 4.20 (1.20) -3.69E-2 (7.1E-3) 1.26E1 (2.05) 1.6E-3 (2.E-4) -2.87E-1 (5.91E-2) -2.9E-3 (7.E-4) -8.33 (9.00E-1) 2.44 (1.46) -1.05E-1 (1.16E-2) 1.70E1 (3.35) Spearmint ‘Spanish’ (Mentha spicata) 2.83 (4.68E-1) 1.56E1 (1.45) 1.66E-2 (6.2E-3) 4.18E1 (6.35) -1.10E1 (3.43) -9.27E-2 (1.04E-2) 2.52E1 (3.30) 1.12 (2.61E-1) 5.37 (8.07E-1) 8.5E-3 (3.2E-3) 2.18E1 (3.54) -3.38 (1.75) -3.33E-2 (5.8E-3) 1.33E1 (1.84) -4.25E-2 (8.2E-3) -2.46E-1 (2.54E-2) -2.E-4 (1.E-4) -7.41E-1 (1.11E-1) 1.71E-1 (5.97E-2) 1.5E-3 (2.E-4) -3.96E-1 (5.79E-2) 5.E-4 (2.E-4) -2.21E-1 (5.68E-2) -2.25E-2 (6.7E-3) -9.45E-2 (2.08E-2) -4.61E-1 (9.12E-2) 9.E-4 (1.E-4) -1.86E-1 (4.74E-2) -7.45E-2 (3.00E-2) -1.42E-2 (7.5E-3) -1.17E-1 (5.03E-2) -5.09 (9.28E-1) -6.92E-1 (1.39E-1) 1.0E-3 (2.E-4) -1.93E-1 (6.35E-2) -8.0E-3 (8.1E-3) -7.73E-2 (2.50E-2) -3.E-4 (1.E-4) -3.07E-1 (1.10E-1) 1.23E-1 (5.89E-2) 2.E-4 (2.E-4) -1.75E-1 (5.70E-2) 7.01E-2 (1.41E-2) 1.05E-2 (3.5E-3) 3.95E-2 (2.36E-2) -2.32 (4.36E-1) 2.22E-1 (6.52E-2) 0.251 0.557 0.140 0.461 0.356 0.400 0.523 0.479 0.179 0.486 0.063 0.409 0.601 0.743 0.661 0.478 0.485 0.102 0.632 Table III-2 (cont’d). Fv/Fm Leaf area (cm2) DMC (g·kg‒1) DMC (g·kg‒1) Leaf mass fraction (fresh) Fresh mass (g) Branch (no.) Height (cm) Fv/Fm Leaf area (cm2) DMC (g·kg‒1) Leaf mass fraction (fresh) Fresh mass (g) 9.04E-1 (1.99E-2) 3.42E2 (2.65E1) 8.70E1 (4.29E1) 1.25E2 (1.32E1) 2.37 (1.72E-1) -2.55E2 (4.99E1) -3.79E1 (6.96) -2.32E2 (2.15E1) 5.40E-1 (9.09E-2) -5.49E2 (9.45E1) 2.69E2 (5.00E1) 2.15 (1.55E-1) -4.18E2 (4.91E1) -3.8E-1 (2.1E-3) -1.11E-1 (1.38E-2) -6.45E-1 (2.55E-1) -1.01E1 (6.04) -1.54 (2.35) -9.47E1 (1.58E1) -1.05E3 (2.91E2) 6.67E1 (2.71) -2.23E2 4.46E-2 (8.51E-1) 2.74E-2 (2.12E-1) 1.61 (1.42) 1.76E2 (2.62E1) 7.16E-1 (2.75E-1) 1.11E1 Sweet basil ‘Nufar’ (Ocimum basilicum) 1.49E-1 (1.42E-1) 3.01E-1 (1.30E-1) 6.81 (8.72E-1) 4.89E1 (1.61E1) Branch (no.) Node (no.) Height (cm) Leaf area (cm2) DMC Fresh mass (g) Table III-2 (cont’d). z Coefficients for model equations were used to generate Figs. III-1‒7 y Blank cells = 0 x Standard error (SE) 1.15E1 (3.81E-2) (4.35E1) -1.96E-1 (2.41) (3.92) 118 40 30 20 10 0 85 80 75 70 65 60 55 120 100 80 60 40 20 0 ) . o n ( h c n a r B ) 1 - g k · g ( C M D ) g ( s s a m h s e r F A C E 22 24 26 30 28 MDT (°C) 32 34 36 4 6 40 30 20 10 0 85 80 75 70 65 60 55 ) . o n ( h c n a r B ) 1 - g k · g ( C M D ) g ( s s a m h s e r F 120 100 80 60 40 20 0 B D F 18 20 12 14 10 8 16 DLI (mol·m-2·d-1) Figure III-1. Mean daily temperature (MDT) effects on purple basil ‘Dark Opal’ (Ocimum basilicum) branch number (A), dry matter concentration (DMC; C), and fresh mass (E), and daily light integral (DLI) effects on branch number (B), DMC (D), and fresh mass (F). Lines represent model predictions, with the coefficients for these models presented in Table III-2. Symbols (means ± SD) represent measured data (A, C, E, n = 27; B, D, F, n = 9). 119 ) m c ( t h g i e H 50 45 40 35 30 25 20 15 10 18 16 14 12 10 DLI (mol·m-2·d-1) 8 6 A m F / v F 0.86 0.84 0.82 0.80 0.78 0.76 0.74 0.72 0.70 36 34 32 30 28 MDT °C 26 24 22 4 18 16 14 12 DLI (mol·m-2·d-1) 10 8 200 180 160 140 120 100 80 60 40 20 0 ) 2 m c ( a e r a f a e L 222426 28 30 MDT °C 32 34 C n o i t c a r f s a m h s e r f f a e L 36 4 6 8 101214161820 L I ( m ol· m -2 ·d-1 ) D 0.90 0.85 0.80 0.75 0.70 0.65 0.60 18 16 14 12 10 DLI (mol·m-2·d-1) 468 B 36 34 32 30 28 26 MDT °C 24 22 6 4 D 22 24 26 2830323436 T ° C M D Figure III-2. Mean daily temperature (MDT) and daily light integral (DLI) effects on purple basil ‘Dark Opal’ (Ocimum basilicum) height (A), maximum quantum yield of dark-adapted leaves (Fv/Fm; B), leaf area (of the four leaves measured; C), and leaf fresh mass fraction (D). Response surfaces represent model predictions. Coefficients for these models are presented in Table III-2. Models are each based on 300 individual measurements. 120 30 28 26 18 16 14 12 DLI (mol·m-2·d-1) 10 8 34 32 30 28 26 MDT °C 24 22 6 4 A B 140 120 100 80 60 40 20 0 -20 -40 -60 e l g n a e u H 18 16 36 34 32 30 28 MDT °C 26 24 22 6 4 14 12 10 DLI (mol·m-2·d-1) 8 * L 48 46 44 42 40 38 36 34 32 30 28 26 18 16 14 12 DLI (mol·m-2·d-1) 10 8 B A 36 34 32 30 28 26 MDT °C 24 22 6 4 Figure III-3. Mean daily temperature (MDT) and daily light integral (DLI) effects on purple basil ‘Dark Opal’ (Ocimum basilicum) hue angle (h°; A) and L* (B). Response surfaces B represent model predictions. Coefficients for these models are presented in Table III-2. Models are each based on 300 individual measurements. g n a 140 120 100 80 60 40 20 0 -20 -40 -60 e l e u H 18 16 14 12 10 DLI (mol·m-2·d-1) 8 6 4 36 34 32 30 28 MDT °C 26 24 22 121 Figure III-4. Mean daily temperature (MDT) and daily light integral (DLI) effects on purple 122 Figure III-4 (cont’d). basil ‘Dark Opal’ (Ocimum basilicum). Replication 1 (A) was harvested on 17 May 2018, and replication 2 (B) was harvested on 27 Nov. 2018, 7 weeks after sowing and 4 weeks after transplant. Images depict representatives of plants measured. 123 ) . o n ( h c n a r B 18 16 14 12 10 8 6 4 2 24 26 28 30 32 34 36 4 MDT °C n o 0.90 0.85 0.80 0.75 0.70 0.65 0.60 0.55 i t c a r f s a m h s e r f f a e L 18 16 14 12 10 DLI (mol·m-2·d-1) 6 8 4 A 60 B ) m c ( t h g i e H 50 40 30 20 10 0 20 18 16 8 10 14 12 DLI (mol·m-2·d-1) 6 36 34 32 30 28 MDT °C 26 24 22 4 18 16 14 12 10 DLI (mol·m-2·d-1) 8 6 C 100 D 80 ) g ( s s a m h s e r F 60 40 20 0 24 26 28 30 32 34 36 4 MDT °C 20 18 10 14 12 16 D LI (m ol·m-2·d-1) 8 6 36 34 32 30 28 MDT °C 26 24 22 Figure III-5. Mean daily temperature (MDT) and daily light integral (DLI) effects on sage ‘Extrakta’ (Salvia officinalis) branch number (A), height (B), leaf fresh mass fraction (C), and leaf fresh mass (D). Response surfaces represent model predictions. Coefficients for these models are presented in Table III-2. Models are each based on 300 individual measurements. 124 240 220 200 180 160 140 120 100 80 200 150 100 50 ) 1 - g k · g ( C M D ) 2 m c ( a e r a f a e L 0 0.86 0.84 0.82 0.80 0.78 0.76 m F v F / 240 220 200 180 160 140 120 100 80 ) 1 - g k · g ( C M D A B 10 12 8 16 DLI (mol·m-2·d-1) 14 18 20 C 4 6 D 22 24 26 28 30 32 34 36 MDT (°C) Figure III-6. Mean daily temperature (MDT) effects on sage ‘Extrakta’ (Salvia officinalis) dry matter concentration (DMC; A), leaf area (C), and maximum quantum yield of dark-adapted leaves (Fv/Fm; D), and daily light integral (DLI) effects on DMC (B). Lines represent model predictions, with the coefficients for these models presented in Table III-2. Symbols (means ± SD) represent measured data (A, C, D, n = 27; B, n = 9). 125 22 20 18 16 14 12 10 8 6 ) . o n ( h c n a r B 20 18 16 14 12 DLI (mol·m-2·d-1) 68 10 4 A ) m c ( 50 40 B t h g i e H 36 34 32 30 28 MDT °C 26 24 22 30 20 10 20 18 16 14 12 DLI (mol·m-2·d-1) 68 10 36 34 32 30 28 MDT °C 26 24 22 4 C ) 2 m c ( 200 180 160 140 120 100 80 60 40 20 2426 a e r a f a e L D 2 1 6 2 8 1 0 2 ol· m - 2 · d- 1 ) 1 4 36 4 6 1 0 8 1 2 L I ( m D 28 M 30 32 DT °C 34 0.86 0.84 0.82 0.80 0.78 0.76 0.74 m F / v F 140 2426 28 30 MDT °C 32 34 2 2 1 2 4 1 0 1 1 6 1 2 8 L I ( m ol· m -2 ·d-1 ) 0 8 36 4 6 D E n o i t c a r f 120 ) 1 - g k · g ( C M D 100 80 22 20 18 16 14 12 DLI (mol·m-2·d-1) 10 68 100 ) g ( s s a m h s e r F 80 60 40 20 0 20 18 16 14 12 DLI (mol·m-2·d-1) 68 10 36 34 32 30 28 MDT °C 26 24 22 G 36 34 32 30 28 MDT °C 26 24 22 4 4 0.75 0.70 0.65 0.60 0.55 0.50 0.45 0.40 s a m h s e r f f a e 20L 18 16 14 12 DLI (mol·m-2·d-1) 68 10 F 36 34 32 30 MDT °C 28 26 24 22 4 Figure III-7. Mean daily temperature (MDT) and daily light integral (DLI) effects on 126 Figure III-7 (cont’d). spearmint ‘Spanish’ (Mentha spicata) branch number (A), height (B), maximum quantum yield of dark-adapted leaves (Fv/Fm; C), leaf area (of the four leaves measured; D), dry matter concentration (DMC; E), leaf fresh mass fraction (F), and fresh mass (G). Response surfaces represent model predictions. Coefficients for these models are presented in Table III-2. Models are each based on 300 individual measurements. 127 20 A ) . o n ( e d o N 9 8 7 6 5 4 3 12 10 8 DLI (mol·m-2·d-1) 6 B 38 36 34 32 30 28 26 24 MDT °C 22 4 38 36 34 32 30 28 26 24 MDT °C 22 15 ) . o n ( h c n a r B 10 5 0 12 10 8 DLI (mol·m-2·d-1) 6 ) m c ( t h g i e H 40 35 30 25 20 15 10 80 ) g ( s s a m h s e r F 60 40 20 0 12 10 8 DLI (mol·m-2·d-1) 6 12 10 DLI (mol·m-2·d-1) 8 6 4 4 4 C ) 2 m c ( a e r a f a e L 38 36 34 32 30 28 26 24 MDT °C 22 700 600 500 400 300 200 100 242628303234 36 38 4 MDT °C 6 D D 14 8 12 10 L I ( m ol· m -2 ·d-1 ) E 38 36 34 32 30 28 26 24 MDT °C 22 Figure III-8. Mean daily temperature (MDT) and daily light integral (DLI) effects on sweet basil ‘Nufar’ (Ocimum basilicum) branch number (A), node number (B), height (C), leaf area 128 Figure III-8 (cont’d). (of the four leaves measured; D), and fresh mass (E). Response surfaces represent model predictions. Coefficients for these models are presented in Table III-2. Models are each based on 300 individual measurements. 129 Sage y = -0.336x + 29.56 Sweet basil A C y = -0.437x + 28.42 Sweet basil Sage B 19 18 17 16 15 14 13 ) 1 - d · 2 - m · l o m ( t p o I L D D 12 13.5 ) 1 - d · 2 - m 13.0 · l o m ( 12.5 t p o 12.0 I L D ) C ° ( t p o T D M ) C ° ( t p o T D M 29 28 27 26 25 24 23 36 35 34 33 32 30.5 30.0 29.5 29.0 28.5 28.0 27.5 ) C ° ( t p o T D M y = 0.566x + 29.23 y = 0.160x + 8.01 30 28 26 Actual MDT (°C) 32 11.5 34 36 Spearmint E 22 24 y = -0.221x + 31.76 4 6 12 10 8 16 Actual DLI (mol·m-2·d-1) 14 18 20 Figure III-9. Predicted optimal mean daily temperature (MDTopt) to achieve the greatest sage ‘Extrakta’ (Salvia officinalis; A), sweet basil ‘Nufar’ (Ocimum basilicum; C), and spearmint ‘Spanish’ (Mentha spicata; E) fresh mass based on actual daily light integral (DLI). Also, predicted DLIopt to achieve the greatest sage (B) and sweet basil (D) fresh mass based on actual MDT. Equations were generated based on a surface regression model (Figures 4D, 6G, and 7E) with model coefficients reported in Table III-2. 130 LITERATURE CITED 131 LITERATURE CITED Adams, S.R., S. Pearson, and P. Hadley, 1997a. The effects of temperature, photoperiod and light integral on the time to flowering of pansy cv. Universal Violet (Viola × wittrockiana Gams.). Annals of Botany, 80(1):107‒112. Adams, S.R., S. Pearson, and P. Hadley, 1997b. An analysis of the effects of temperature and light integral on the vegetative growth of pansy cv. Universal Violet (Viola× wittrockiana Gams.). Annals of Botany, 79(3):219‒225. Armitage, A.M., W.H., Carlson, and J.A. Flore, 1981. The effect of temperature and quantum flux density on the morphology, physiology, and flowering of hybrid geraniums. Journal of the American Society for Horticultural Science, 106(5):643‒647. Blanchard, M.G., E.S. Runkle, and P.R. Fisher, 2011. Modeling plant morphology and development of petunia in response to temperature and photosynthetic daily light integral. Scientia Horticulturae, 129(2):313‒320. Boese, S.R. and N.P. Huner, 1990. Effect of growth temperature and temperature shifts on spinach leaf morphology and photosynthesis. Plant Physiology, 94(4):1830‒1836. Busch, G., E. Bayer, A. Gunarathne, S. Hölker, S. Iweala, K. Jürkenbeck, D. Lemken, C. Mehlhose, M. Ohlau, A. Risius, and C. Rubach, 2020. Purchasing and nutritional behavior and resilience of the nutritional system from a population perspective: Results of a study during the Corona pandemic in April 2020 (No. 2003). Georg-August University of Göttingen, Department of Agricultural Economics and Rural Development. Catley, J.L. and I.R. Brooking, 1996. Temperature and light influence growth and flower production in heliconia ‘Golden Torch'. HortScience, 31(2):213‒217. Chang, X., P. Alderson, and C. Wright, 2005. Effect of temperature integration on the growth and volatile oil content of basil (Ocimum basilicum L.). The Journal of Horticultural Science and Biotechnology, 80:593‒598. Chermnykh, L.N. and A.A. Kosobrukhov, 1987. Effect of environmental factors on optimum temperature and photosynthetic intensity of plants adapted to various conditions. Biotronics, 16:1‒11. de Koning, A.N.M., 1994. Development and dry matter distribution in glasshouse tomato: A quantitative approach. Wageningen Agricultural University, The Netherlands. PhD Dissertation. Dou, H., G. Niu, M. Gu, and J.G. Masabni, 2018. Responses of sweet basil to different daily light integrals in photosynthesis, morphology, yield, and nutritional quality. HortScience, 53:496–503. 132 Enoch, H.Z. and R.G. Hurd, 1977. Effect of light intensity, carbon dioxide concentration, and leaf temperature on gas exchange of spray carnation plants. Journal of Experimental Botany, 28(1):84–95. Faust, J.E. and R.D. Heins, 1998. Modeling shoot-tip temperature in the greenhouse environment. Journal of the American Society for Horticultural Science, 123(2):208–214. Faust, J.E., V. Holcombe, N.C. Rajapakse, and D.R. Layne, 2005. The effect of daily light integral on bedding plant growth and flowering. HortScience, 40(3):645–649. Frąszczak, B. and M. Knaflewski, 2009. Effect of light conditions and temperature on fresh yield of some spice plants grown in containers. Vegetable Crops Research Bulletin, 71:59–67. Gomez, C., C.J. Currey, R.W. Dickson, H. Kim, R. Hernandez, N.C. Sabeh, R.E. Raudales, R.G. Brumfield, A. Laury-Shaw, A.K. Wilke, R.G. Lopez, and S.E. Burnett, 2019. Controlled environment food production for urban agriculture, HortScience, 54:1448–1458. Karlsson, M.G., R.D. Heins, J.O. Gerberick, and M.E. Hackmann, 1991. Temperature driven leaf unfolding rate in Hibiscus rosa-sinensis. Scientia Horticulturae, 45:323–331. Kim, S., G. Hwang, S. Lee, J.Y. Zhu, I. Paik, T.T. Nguyen, J. Kim, and E. Oh, 2017. High ambient temperature represses anthocyanin biosynthesis through degradation of HY5. Frontiers in Plant Science, 8:1787. Kolodinsky, J., M. Sitaker, L. Chase, D. Smith, and W. Wang, 2020. Food systems disruptions. Journal of Agriculture, Food Systems, and Community Development, 9:1–4. Korczynski, P.C., J. Logan, and J.E. Faust, 2002. Mapping monthly distribution of daily light integrals across the contiguous United States. HortTechnology, 12(1):12–16. Lasseigne, F.T., S.L. Warren, F.A. Blazich, and T.G. Ranney, 2007. Day/night temperature affects growth and photosynthesis of cultivated Salvia taxa. Journal of the American Society for Horticultural Sciences, 132:492–500. Leyva, A., J.A. Jarillo, J. Salinas, and J.M. Martinez-Zapater, 1995. Low temperature induces the accumulation of phenylalanine ammonia-lyase and chalcone synthase mRNAs of Arabidopsis thaliana in a light-dependent manner. Plant Physiology, 108:39–46. Lin‐Wang, K.U.I., D. Micheletti, J. Palmer, R. Volz, L. Lozano, R. Espley, R.P. Hellens, D. Chagne, D.D. Rowan, M. Troggio, and I. Iglesias, 2011. High temperature reduces apple fruit colour via modulation of the anthocyanin regulatory complex. Plant, Cell, and Environment, 34:1176–1190. Litvin, A.G., 2019. Quantifying the effects of light quantity and quality on culinary herb physiology. Iowa State University, Ames, PhD Dissertation. McGuire, R.G., 1992. Reporting of objective color measurements. HortScience, 27(12):1254– 1255. 133 McLellan, M.R., L.R. Lind, and R.W. Kime, 1995. Hue angle determinations and statistical analysis for multiquadrant Hunter L, a, b data. Journal of Food Quality, 18(3):235–240. Moccaldi, L.A. and E.S. Runkle, 2007. Modeling the effects of temperature and photosynthetic daily light integral on growth and flowering of Salvia splendens and Tagetes patula, Journal of the American Society for Horticultural Science, 132:283–288. Mori, K., N. Goto-Yamamoto, M. Kitayama, and K. Hashizume, 2007. Loss of anthocyanins in red-wine grape under high temperature. Journal of Experimental Botany, 58:1935–1945. Mortensen, L.M., 2014. The effect of air temperature on growth of eight herb species. American Journal of Plant Sciences, 5:1542–1546. Movahed, N., C. Pastore, A. Cellini, G. Allegro, G. Valentini, S. Zenoni, E. Cavallini, E. D’Incà, G.B. Tornielli, and I. Filippetti, 2016. The grapevine VviPrx31 peroxidase as a candidate gene involved in anthocyanin degradation in ripening berries under high temperature. Journal of Plant Research, 129:513–526. Niu, G., R.D. Heins, A.C. Cameron, and W.H. Carlson, 2001. Day and night temperatures, daily light integral, and CO2 enrichment affect growth and flower development of Campanula carpatica ‘Blue Clips’. Scientia horticulturae, 87(1–2):93–105. Oh, W., J. Kim, Y.H. Kim, I.J. Lee, and K.S. Kim, 2015. Shoot elongation and gibberellin contents in Cyclamen persicum are influenced by temperature and light intensity. Horticulture, Environment, and Biotechnology, 56(6):762–768. Phippen, W.B. and J.E. Simon, 1998. Anthocyanins in basil. Journal of Agricultural Food Chemistry, 46:1734–1738. Pogany, D., C.L. Bell, and E.R. Kirch, 1968. Composition of oil of sweet basil (Ocimum basilicum L.) obtained from plants grown at different temperatures. Perfumery Essential Oil Record, 59:858–865. Poorter, H., K.J. Niklas, P.B. Reich, J. Oleksyn, P. Poot, and L. Mommer, 2012. Biomass allocation to leaves, stems and roots: meta‐analyses of interspecific variation and environmental control. New Phytologist, 193(1):30–50. Pramuk, L.A. and E.S. Runkle, 2005. Modeling growth and development of celosia and impatiens in response to temperature and photosynthetic daily light integral. Journal of the American Society for Horticultural Science, 130(6):813–818. Putievsky, E., 1983. Temperature and daylength influences on the growth and germination of sweet basil and oregano. Journal of Horticultural Science, 58:583–587. Rawson, H.M. and J.H. Hindmarsh, 1982. Effects of temperature on leaf expansion in sunflower. Functional Plant Biology, 9(2):209–219. 134 Rehman, R.N.U., Y. You, L. Zhang, B.D. Goudia, A.R. Khan, P. Li, and F. Ma, 2017. High temperature induced anthocyanin inhibition and active degradation in Malus profusion. Frontiers Plant Science, 8:1401. Sage, R.F. and D.S. Kubien, 2007. The temperature response of C3 and C4 photosynthesis. Plant, Cell, and Environment, 30:1086–1106. U.S. Department of Agriculture (USDA) – National Agriculture Statistics Service. 2010. 2009 U.S. Census of Horticultural Specialties. 13 Mar. 2019. . U.S. Department of Agriculture (USDA) – National Agriculture Statistics Service. 2015. 2014 U.S. Census of Horticulture Specialties. 13 Mar. 2019. . Walters, K.J. and C.J. Currey, 2015. Hydroponic greenhouse basil production: Comparing systems and cultivars. HortTechnology, 25:645−650. Walters, K.J. and C.J. Currey, 2019. Growth and development of basil species in response to temperature. HortScience, 54:1915−1920. White, J.W. and I.J. Warrington, 1984. Growth and development responses of geranium to temperature, light integral, CO2, and chlormequat. Journal of the American Society for Horticultural Science, 109(5):728−735. 135 CHAPTER 4 LEVERAGING CONTROLLED-ENVIRONMENT AGRICULTURE TO INCREASE KEY BASIL TERPENOID AND PHENYLPROPANOID CONCENTRATIONS: THE EFFECTS OF RADIATION INTENSITY ON CONSUMER PREFERENCE 136 Leveraging controlled-environment agriculture to increase key basil terpenoid and phenylpropanoid concentrations: The effects of radiation intensity on consumer preference Kellie J. Walters and Roberto G. Lopez Acknowledgements We gratefully acknowledge Sean Tarr, Nate DuRussel, Bridget Behe, Randy Beaudry, Cassandra Johnny, and Alex Renny for assistance, Fluence Bioengineering for LEDs, JR Peters for fertilizer, Grodan for substrate, Hydrofarm for hydroponic production systems, and the Michigan State University Research Technology Support Facility Mass Spectrometry and Metabolomics Core for GCMS access and method guidance. This work was supported by Michigan State University AgBioResearch (including Project GREEEN GR19-019), the USDA National Institute of Food and Agriculture (Hatch project MICL02472), and The Fred C. Gloeckner Foundation. The use of trade names in this publication does not imply endorsement by Michigan State University of products named nor criticism of similar ones not mentioned. 137 Additional Index Words: 1,8 cineole, daily light integral, estragole, eucalyptol, eugenol, linalool, methyl chavicol, Ocimum basilicum, sensory panel Abstract. Radiation intensity, specifically photosynthetic photon flux density (PPFD) plays a large role in the growth and development of plants. However, altering the radiation intensity in controlled environments (CE) can also influence volatile organic compound (VOC) biosynthetic pathways, including those of terpenoids and phenylpropanoids. In turn, the concentrations of these compounds can have a profound effect on flavor and sensory attributes of fresh culinary herbs. Since sweet basil (Ocimum basilicum) is a popular culinary herb, our objectives were to 1) determine the extent radiation intensity and carbon dioxide (CO2) concentration influence key basil seedling terpenoid and phenylpropanoid concentrations; 2) determine if differences in key phenylpropanoid and terpenoid concentrations due to radiation intensity and CO2 concentration influence consumer preference; and 3) characterize consumer preferences to better inform production and marketing strategies. Seeds of basil ‘Nufar’ were sown and placed in a growth chamber with CO2 concentrations of 500 or 1,000 µmol·mol‒1 under radiation intensities of 100, 200, 400, or 600 µmol·m–2·s–1 PPFD with a 16-h photoperiod to create daily light integrals (DLIs) of 6, 12, 23, and 35 mol·m–2·d–1. After two weeks, leaves were harvested for evaluation. To determine the concentrations of the terpenoids 1,8 cineole and linalool, and the phenylpropanoids eugenol and methyl chavicol, gas chromatography mass spectrometry (GCMS) analysis was conducted. Consumer sensory panel evaluations were conducted to quantify preferences through Likert scale, open-ended quality attribute, and sensory questions. Overall, increasing radiation intensity from 100 to 600 µmol·m‒2·s‒1 increased 1,8 cineole, 138 linalool, and eugenol concentrations 2.4-, 8.8-, and 3.3-fold, respectively, while CO2 concentration did not influence VOCs. Contrary to our hypothesis, increased VOC concentrations were not correlated with consumer preference. However, overall liking was correlated with aftertaste and flavor. The conclusion that consumer preference is dependent upon flavor can be drawn. However, increasing VOC concentrations to increase flavor did not improve flavor preference. Many consumer sensory preference characteristics (favorable preference for aftertaste, bitterness/sweetness, color, flavor, overall liking, and texture) were correlated with basil grown under a radiation intensity of 200 µmol·m‒2·s‒1. This led us to determine that consumers prefer to detect the characteristic basil flavor made up of 1,8 cineole, eugenol, and linalool, that was not as prevalent in basil grown under 100 µmol·m‒2·s‒1, but too high in basil grown under 400 and 600 µmol·m‒2·s‒1, which led to lower consumer preference. Introduction Increased demand for a year-round supply of locally grown produce in urban markets, interest in climate change resilience, and the mitigation of food deserts has spurred interest in indoor controlled-environment agriculture (CEA; Gomez et al., 2009; Goodman and Minner, 2018; Metz et al., 2007; Reynolds and Cohen, 2016; Webber and Matthews, 2008). However, the high capital and operating costs of indoor CEA can cause questionable economic feasibility of most food crops (Eaves and Eaves, 2017; Hoops et al, 2018; Zeidler et al., 2013). Therefore, the production of high-value and high-quality specialty crops such as leafy greens has prevailed in CEA, with growers indicating a need for research on manipulating the growing environment to improve crop flavor (Goodman and Minner, 2018; Walters et al., 2020). 139 A predominant high-value specialty crop that can vary greatly in quality is fresh cut sweet basil (Ocimum basilicum). Basil varies not only in visual quality, but in flavor caused by variations in the concentration and ratios of volatile organic compounds (VOCs). Different basil species and cultivars have varying chemotypes, which can be characterized by distinct dominant compounds or ratios of compounds, and therefore, different overall flavors. For example, lemon basil cultivars including ‘Sweet Dani’ contain high concentrations of citral (terpenoid), giving them a lemony flavor and aroma (Morales and Simon, 1997). Other cultivars have higher concentrations of linalool (terpenoid), methyl chavicol or eugenol (phenylpropanoids), or more than one major compound, leading to wide variation in basil flavor and aroma (Gang et al., 2001; Simon et al., 1999). Many of the VOCs contributing to basil flavor are terpenoids or phenylpropanoids. The biosynthetic pathways for these two compound groups have differing rate-limiting steps and regulatory mechanisms. For example, terpenoid concentration is correlated with terpene synthase activity but negatively correlated with phenylpropanoid concentration and phenylalanine ammonia lyase (PAL) activity (Iijima et al., 2004). Additionally, increased energy and substrate availability generally promotes VOC biosynthesis. Therefore, we hypothesized that increased concentrations of both terpenoids and phenylpropanoids could be achieved by increasing radiation intensity during production, and thereby, increasing substrate and energy availability. Additionally, we hypothesized the ratios between compounds would differ based on the radiation intensity provided due to the differing biosynthetic pathways. Based on our preliminary analysis to determine the major compounds present in sweet basil ‘Nufar’, two phenylpropanoids and two terpenoids were chosen for analysis (data not shown). Eugenol, a major phenylpropanoid contributing to the clove-like flavor and aroma of 140 basil, is the major volatile oil in cloves (Syzygium aromaticum; Santos et al., 2009). It also has antibacterial, antifungal, and anti-herbivory characteristics (Karapinar and Aktug, 1987; Moleyar and Narasimham, 1992; Sangwan et al., 1990; Singh et al., 2015; Sisk et al., 1996). The second phenylpropanoid, methyl chavicol (estragole), contributes to the anise-like aroma and flavor characteristic of basil (Simon et al., 1999). Linalool, a monoterpenoid, can be described as having an aroma and flavor of floral or spicy (Arena et al., 2006) or reminiscent of the cereal Fruit Loops®. Linalool also has antibacterial, antifungal, and insecticidal activity (Juliani et al., 2004). The second major monoterpenoid is 1,8 cineole (eucalyptol). With an aroma and flavor analogous to eucalyptus (Eucalyptus globulus), whose essential oil profile contains 70% to 80% 1,8 cineole (De Vincenzi et al., 2002), the compound also has insecticidal activity (Shaaya et al., 1991). Researchers have investigated the relationship between radiation intensity and/or the daily light integral (DLI) and secondary metabolite concentrations. In general, overall VOC concentration increases as DLI increases. However, trends differ among individual compounds; this has been demonstrated in many culinary herbs including basil, dill (Anethum graveolens), sage (Salvia officinalis), and thyme (Thymus vulgaris) (Chang et al., 2007; Hälvä et al., 1992; Kumar et al., 2013; Li et al., 1996). In basil, linalool and eugenol concentrations increased ~3- fold and ~4-fold, respectively, while methyl eugenol decreased by ~80% and 1,8-cineole was unaffected as DLI increased from 5 to 25 mol·m‒2·d‒1 (Chang et al., 2008). Dou et al. (2018) determined that increasing the DLI from 9 to 18 mol·m–2·d–1 not only increased basil fresh mass, net photosynthesis, and leaf area and thickness, but also increased anthocyanin, phenolic, and flavonoid concentrations. 141 Though researchers have investigated the influence of DLI on secondary metabolite accumulation, there are limited data on how these aroma and flavor profile changes affect consumer preference. If a production goal is to produce a premium quality product with improved flavor characteristics, connecting the concentration of VOCs that contribute to flavor with consumer preferences could improve crop flavor and consumer demand. Sensory analysis panels have been conducted to determine perceived differences in basil aroma due to radiation source, radiation quality, and temperature during production (Chang et al., 2007; Seely, 2017), to determine perceived differences in drying methods (Calin-Sanchez et al., 2012; Diaz-Maroto et al., 2004), and to characterize basil cultivars (Bernhardt et al., 2015; D’Antuono et al., 2007; de Costa et al., 2014; Tangpao et al., 2018). Although some of these studies connect production practices with consumer preference, recommendations based on radiation intensity are needed. While increasing crop quality is one method to improve economic feasibility, high- density planting is another strategy utilized to mitigate the high input costs of CEA by reducing the cost per plant. Plant density is generally greatest during the seedling production stage versus the finishing stage(s). Therefore, if greater inputs are used during the seedling stage, the cost can be spread across more plants. To leverage the environmental control capabilities of CEA to improve VOC concentrations and ratios most efficiently, our objectives were to: 1) determine the extent radiation intensity influences key sweet basil seedling terpenoid and phenylpropanoid concentrations; 2) determine if differences in key phenylpropanoid and terpenoid concentrations due to radiation intensity influence consumer preference; and 3) characterize consumer preferences to better inform production and marketing strategies. We hypothesized that VOC concentration would increase as the radiation intensity increased. We also postulated that consumers would prefer basil with higher VOC concentrations and a more intense flavor. 142 Materials and Methods Seedling production. Sweet basil ‘Nufar’ (Johnny’s Selected Seeds, Fairfield, ME) was selected based on yield data from Walters and Currey (2015). Seeds were sown two per cell in stone wool cubes (2.5 × 2.5 × 4 cm, AO plug; Grodan, Roermond, Netherlands), with 200-cell flats placed in one of two walk-in growth chambers (Hotpack environmental room UWP 2614-3; SP Scientific, Warminster, PA). Seeds and seedlings were irrigated overhead daily with deionized water supplemented with 12N–1.76P–13.44K water-soluble fertilizer providing (mg·L–1) 100 nitrogen, 15 phosphorus, 112 potassium, 58 calcium, 17 magnesium, 2 sulfur, 1.4 iron, 0.5 zinc, 0.4 copper and manganese, and 0.1 boron and molybdenum (RO Hydro FeED; JR Peters, Inc., Allentown, PA) and magnesium sulfate (MgSO4) providing (mg·L–1) 15 magnesium and 20 sulfur. Seedlings were thinned to one seedling per cell approximately one week after sowing. The air temperature setpoint was 23 °C; air temperature was measured by a resistance temperature detector (Platinum RTD RBBJL-GW05A-00-M 36B; SensorTec, Inc., Fort Wayne, IN) every 5 s and logged by a C6 controller (Environmental Growth Chambers, Chagrin Falls, OH) with values reported in Table IV-1. Substrate temperature was measured with a thermistor (ST-100; Apogee Instruments, Logan, UT), leaf temperature was measured with an infrared thermocouple (OS36-01-T-80F; Omega Engineering, INC. Norwalk, CT), and photosynthetic photon flux density (PPFD) was monitored with a quantum sensor (LI-190R Quantum Sensor; LI-COR Biosciences, Lincoln, NE) every 15 s with means logged every hour by a CR-1000 datalogger (Campbell Scientific, Logan, UT). Target carbon dioxide concentrations of 500 and 1000 µmol·mol–1 were maintained by injecting compressed CO2 to increase concentrations and 143 scrub CO2 using soda lime scrubber (Environmental Growth Chambers) to decrease concentrations. Concentrations were measured with a CO2 sensor (GM86P; Vaisala, Helsinki, Finland) and logged by a C6 Controller (Environmental Growth Chambers) every 5 s (Table IV- 1). LEDs (Ray66 Indoor PhysioSpec; Fluence Bioengineering, Austin, TX) provided 20:40:40 blue:green:red radiation ratios (%), a red:far-red ratio of 13:1, and target PPFDs of 100, 200, 400, or 600 µmol·m‒2·s‒1 for 16 h to create DLIs of 6, 12, 23, or 35 mol·m‒2·d‒1, respectively (Fig. IV-1). Fixture density and hanging height were adjusted to achieve target radiation intensities. Radiation intensity and spectrum were measured at four corners and in the center of the seedling flat with a spectroradiometer (PS-200; StellarNet, Inc., Tampa, FL) to quantify the intensities and spectrum (Fig. IV-1) across the growing area with the variation reported in Table IV-1. VOC data collection and analysis. Two weeks after sowing the two most recent, fully mature leaves of five plants per treatment, per replication, were detached, frozen, and stored at ‒20 °C until gas chromatography mass spectrometry (GCMS) analysis. In a method derived from Schilmiller et al. (2010), tissue was ground in liquid nitrogen and an aliquot was placed in a 1.5 mL microcentrifuge tube containing 500 µL of methyl tert-butyl ether (MTBE) with 10 ng·µL–1 of tetradecane internal standard and gently rocked for 3 minutes. Samples were centrifuged to pelletize the tissue and 150 µL of supernatant was transferred to gas chromatography auto sampler vials. Samples were analyzed using an Agilent 7890A GC and single quadrupole MS with 5975C inert XL mass spectrometry detector (Agilent, Santa Clara, CA). Standards were utilized to identify metabolites in addition to m/z values in the ChemStation database. Peak areas were integrated using 144 MassLynx V4.1 QuanLynx software (Waters Corporation, Milford, MA). The compound concentrations were normalized to the sample tetradecane internal standard and leaf dry weight, then quantified using the standard calibration curves of 1,8 cineole, eugenol, linalool, and methyl chavicol with a tetradecane internal standard (MilliporeSigma; St. Louis, MO). Sensory analysis. The protocol for the sensory analysis portion of this study was approved by the Institutional Review Board of Michigan State University (MSU; East Lansing, MI, USA; STUDY000001369; Appendix C). The experimental procedure was thoroughly explained to all participants and a written informed consent was obtained from each prior to participation. Two consumer sensory panels were conducted using the Sensory Evaluation Laboratory in the Department of Food Science and Human Nutrition at MSU. The first compared four radiation intensity treatments, 100, 200, 400, and 600 µmol·m‒2·s‒1 with plants grown in 500 µmol·mol‒1 CO2. The second compared two CO2 concentrations, 500 and 1,000 µmol·mol‒1, and two radiation intensities, 200 and 400 µmol·m‒2·s‒1. Ninety and 98 participants, respectively, were recruited through the MSU Communication Arts and Sciences Paid Research Pool and screened prior to participation to ensure they had consumed basil in the past four months. Leaf samples were harvested 1 to 4 h prior to sampling to ensure freshness. Individual leaves were removed and rinsed in deionized water. Samples were dried with a salad spinner and/or through air-drying. Depending on leaf size, four to eight leaves per sample were placed in cups. Participants sat in a booth containing a roll up pass-through door with a computer above the door, fluorescent lighting, and positive air pressure. Participants answered sensory evaluation and demographic questions through the Sensory Information Management System (SIMS 2000 version 6.0, Berkeley Heights, NJ) on the booth computer. Provided with water and saltine 145 crackers, participants were served samples individually in a random order with 3-digit blinding codes. Upon receipt of a sample, participants answered questions (Appendix B) on a 9-point Likert scale ranging from dislike extremely to like extremely to describe how much they liked the sample over all, the aftertaste, appearance, aroma, flavor, texture, and leaf size; they rated the level of bitterness or sweetness; described what they liked and disliked about the sample; and shared any additional comments they had. Once a sample was evaluated, the sample was removed and another sample was provided, prompting the same questions. Upon completion of the sensory evaluation, participants provided demographic information. Statistical design and analysis. This experiment was arranged in a split-plot design with two CO2 concentrations (two growth chambers) as the main factor and four radiation intensities as the sub factor. The experiment was completed twice in time for GCMS analysis (reps 1 and 2) and once in time for consumer sensory analysis (rep 2). Analysis of variance was performed on VOC and Likert data using JMP (version 12.0.1, SAS Institute Inc., Cary, NC). Linear and quadratic regression analyses were conducted on VOC data using SigmaPlot (version 11.0, Systat Software Inc., San Jose, CA) and Tukey’s honestly significant difference test (P< 0.05) was conducted on Likert data using JMP. Chi2 test (P< 0.05) was conducted to analyze word frequency using WordStat (version 8, Provalis Research, Montreal, Canada). Data were transformed to account for differences in values, variation, and sample size [(sample value – parameter average)/parameter SD], then principal component analysis was conducted (using JMP). Biplots were created combining principal component analysis with loading plots. Factor analysis was conducted to determine the significant factors 146 based on a rotated factor loading value of > 0.7. Correlations were determined by Pearson’s correlation coefficient at P <0.05. Results Seedling volatile organic compound concentrations. CO2 concentration did not influence VOC concentration (data not shown), so VOC data were pooled across both CO2 concentrations for each radiation intensity. The concentrations of both measured terpenoids increased linearly as radiation intensity increased, but the extent varied. Increasing the radiation intensity from 100 to 600 µmol·m‒2·s‒1 increased 1,8 cineole concentration from 450 to 1,510 ng·mg‒1 dry mass (DM; 2.4-fold) while linalool concentration increased from 67 to 655 ng·mg‒1 DM (8.8-fold; Figs. IV-2A and B). The relationship between eugenol concentration and radiation intensity was quadratic. A 56% (283 ng·mg‒1 DM) decrease in concentration occurred as radiation intensity increased from 100 to 200 µmol·m‒2·s‒1, with an 8.7-fold (1952 ng·mg‒1 DM) increase as radiation intensity further increased from 200 to 600 µmol·m‒2·s‒1 (Fig. IV-2C). Average methyl chavicol concentration tended to increase as radiation intensity increased, however, due to large variations in concentration, differences between means were not significant (Fig. IV-2D). Sensory panel. The mean age of consumer panelists was 30.4 years, with an average of 1.2 adults and 0.4 minors in the household. Sixty-nine percent of panelists were female, 30% were male, 68% were Caucasian, and 19% were Asian. Average household income was $50,000, and 71% had at least a 4-year post-secondary education. Overall, consumers preferred basil grown under a radiation intensity of 200 compared to 600 µmol·m‒2·s‒1 (Fig. IV-3). Flavor preference followed the same trend as overall preference, 147 with consumers liking the flavor of basil grown under 200 more than 600 µmol·m‒2·s‒1 (Fig. IV- 4A). Similarly, the aftertaste of plants grown under 200 µmol·m‒2·s‒1 was preferred over those grown under 400 or 600 µmol·m‒2·s‒1 (Fig. IV-4B). Basil grown under 200 µmol·m‒2·s‒1 was the least bitter. Increasing the radiation intensity to 400 or 600 µmol·m‒2·s‒1 resulted in leaves that consumers rated as slightly bitter (Fig. IV-4D). The aroma of plants grown under 100 µmol·m‒ 2·s‒1 was the least-preferred, with likeability increasing as radiation intensity increased to 200 µmol·m‒2·s‒1 then plateaued (Fig. IV-4C). Visually, the appearance of basil grown under 400 µmol·m‒2·s‒1 was preferred to those grown under 100 µmol·m‒2·s‒1 (Figs IV-5 and IV-6A). The color of basil grown under 100, 200, or 400 µmol·m‒2·s‒1 was preferred more than the color of plants grown under 600 µmol·m‒2·s‒1 (Figs. IV-5 and IV-6B). The size of leaves produced under higher radiation intensities (400 or 600 µmol·m‒2·s‒1) was preferred, with likeability decreasing as radiation intensity decreased (Figs. IV-5 and IV-6C). Leaf texture preference followed a similar trend to overall liking in that panelists preferred plants grown under 200 µmol·m‒2·s‒1 more than those grown under 600 µmol·m‒2·s‒1 (Fig. 6 IV-D). Based on word frequency, “bitter” was associated more with basil grown under higher radiation intensities (400 or 600 µmol·m‒2·s‒1; Fig. IV-7). The words “brown”, “chewy”, “odd”, “spicier”, and “wilted” were most strongly associated with plants grown under 600 µmol·m‒2·s‒1 that, after rinsed for panel analysis, exhibited symptoms of leaf damage (Fig IV-5.). The words “mouth”, “quickly”, “reminds”, and “yellow” were most frequently used to comment on the 400 µmol·m‒2·s‒1 grown plants. In comments regarding 100 and 200 µmol·m‒2·s‒1 grown plants, “small” was frequently used; 100 µmol·m‒2·s‒1 grown plants were also described with the words 148 “buy”, “enjoyable”, “spice”, “subtle”, and “tiny”, and 200 µmol·m‒2·s‒1 grown plants were described as “vibrant”. In the second sensory panel, there were no differences in consumer preference due to CO2 concentration during production (data not shown). However, they confirmed the differences noted by the first sensory panel between basil grown under 200 or 400 µmol·m‒2·s‒1. For example, basil grown under 200 µmol·m‒2·s‒1 was less bitter than basil grown under 400 µmol·m‒2·s‒1 (data not shown). Comparing concentrations to sensory panel. A principal component analysis comparison of VOCs and consumer sensory preferences, represented by a biplot including basil samples grown under 100, 200, 400, or 600 µmol·m‒2·s‒1, determined that two components accounted for 92% of the total data variability (Fig. IV-8). Component 1 separates basil grown under 200 µmol·m‒2·s‒1 from those grown under 600 µmol·m‒2·s‒1. The positive discriminating factors for component 1 are aftertaste, bitterness/sweetness, color, flavor, overall liking, and texture while the negative factors are 1,8 cineole, eugenol, and linalool concentrations. Component 2 separates basil grown under 100 µmol·m‒2·s‒1 from those grown under 200, 400, and 600 µmol·m‒2·s‒1. The positive discriminating factors are aroma, appearance, methyl chavicol concentration, and leaf size and there are no negative discriminating factors. While 1,8 cineole, linalool, and eugenol concentrations were positively correlated with each other, they were negatively correlated with color, and linalool was negatively correlated with bitterness/sweetness (Fig. IV-9). Additionally, aftertaste, flavor, and overall liking were positively correlated (Fig. IV-9). 149 Discussion Sweet basil is consumed primarily for its distinct flavor. However, that flavor is highly variable based on cultivar genetics and the growing environment. An advantage to CEA production is that the growing environment can be monitored and adjusted to produce a consistent and potentially high-quality crop year-round. A major factor contributing to quality is the concentration of VOCs that contribute to crop flavor. Terpenoids. The linear increase in linalool and 1,8 cineole concentrations is congruent with previous research. Similar to Chang et al. (2008) who reported a ~4-fold increase in linalool concentration as DLI increased from 5 to 25 mol·m‒2·d‒1, we also found a nearly 4-fold increase as DLI increased from 6 mol·m‒2·d‒1 (radiation intensity of 100 µmol·m‒2·s‒1) to 23 mol·m‒2·d‒1 (400 µmol·m‒2·s‒1) (Fig. IV-2B). This linear increase was likely due to increasing radiation intensity increasing substrate availability (Chang et al., 2008). Although substrate availability may be the main limiting factor in monoterpenoid biosynthesis, other regulatory enzymes may also play a role (Munoz-Bertomeu et al., 2006). Biosynthesis of monoterpenoids takes place in the plastid, with the precursors, isopentenyl diphosphate and dimethylallyl diphosphate, synthesized through the methylerythritol 4-phosphate (MEP) pathway. The MEP pathway begins with the first and main monoterpenoid biosynthesis rate-limiting step, the condensation of pyruvate and glyceraldehyde 3-phosphate catalyzed by 1-deoxy-D-xyluloase 5-phosphate synthase (DXS) to create 1-deoxy-D-xylulose 5-phosphate (DXP; Rodriguez-Concepcion and Boronat, 2015; Simpson et al., 2016; Wright et al., 2014). The DXS encoding gene transcripts can increase 2- to 9-fold in the presence of light (Kim et al., 2005). Additionally, the level of DXS is directly positively correlated to the concentration of terpenoid and terpenoid products (Enfissi et al., 150 2005; Estevez et al., 2000; Gong et al., 2006). Beyond the MEP pathway, a series of terpene synthases catalyze reactions to synthesize our compounds of interest, linalool and 1,8 cineole, through carbocationic reactions. Terpenoid concentration is directly correlated with terpene synthase activity (Iijima et al., 2004). Phenylpropanoids. In general, phenylpropanoid (eugenol and methyl chavicol) concentration increased as radiation intensity increased. However, as radiation intensity increased from 6 to 12 mol·m‒2·d‒1 (100 to 200 µmol·m‒2·s‒1), eugenol concentration decreased by 56% (Fig. IV-2), whereas Chang et al. (2008) reported an increase in relative eugenol content as the DLI increased from 5 to 11 mol·m‒2·d‒1 and a reduction from 11 to 14 mol·m‒2·d‒1. Though the dip in relative eugenol content as radiation intensity increased occurred at different intensities (or DLIs), the dip still remains. However, similar to our results, Chang (2008) reported that eugenol concentration increased ~3-fold as DLI increased from 5 to 25 mol·m‒2·d‒1; we found an over 3-fold increase when increasing the DLI from 6 to 23 mol·m‒2·d‒1 (Fig. IV-2). The concentration of phenylpropanoids is largely dependent upon enzymatic concentration and activity, including PAL (Iijima et al., 2004). In a study comparing basil cultivar secondary metabolite variation, the cultivar with the highest phenylpropanoid concentrations (breeding line EMX) had PAL activity 2.8 times higher than the cultivar with the lowest phenylpropanoid concentration (‘Sweet Dani’; Iijima et al., 2004). Also, the cultivar with phenylpropanoid concentration in between (breeding line SW) had intermediate PAL activity. They found several related sequences encoding PAL and determined PAL gene transcript levels mirrored phenylpropanoid concentration across the breeding lines or cultivars evaluated (Iijima et al., 2004). In basil, researchers have also reported increased levels of other compounds 151 produced downstream of PAL in response to increased DLI including anthocyanins, total phenolics, and flavonoids (Dou et al., 2018). PAL catalyzes the first committed step in the phenylpropanoid pathway, the removal of ammonia (NH3) from phenylalanine to create cinnamic acid. PAL is a highly regulated enzyme whose activity is dependent upon many factors. It is upregulated by low or high temperature (Christie et al., 1994; Leyva et al., 1995; Olsen et al., 2008; Rivero et al., 2001); drought (Falahi et al., 2018; Oh et al., 2009); mineral nutrition, specifically low nitrogen (Fritz et al., 2006; Olsen et al., 2008); ultraviolet radiation (Kuhn et al., 1984); pathogens (Haard and Wasserman, 1976; Kuhn et al., 1984); plant age and lignin production (Anterola and Lewis, 2002); tissue wounding (Haard and Wasserman, 1976; Wong et al., 1974); ethylene (Riov et al., 1969); and radiation quality, where a mix of R and B radiation increased PAL activity more than B or R radiation alone (Heo et al., 2012). Additionally, specific phenylpropanoids are also regulated further downstream. The decrease in methyl chavicol concentration as leaves mature is caused partially by a reduction in chavicol O-methyltransferase (CVOMT) and eugenol O-methyltransferase (EOMT) activity (Deschamps et al., 2005). Differential regulation may have led to the large variation in methyl chavicol concentrations in this seed-propagated cultivar. Compound sensitivity. Preference or dislike for foods can be due to compounds present in very small concentrations (Dris and Jain, 2004). Human olfactory detection and recognition thresholds are the minimum concentrations of a compound that panelists can detect the presence of and recognize the compound, respectively (Patton and Josephson, 1957). Olfactory thresholds can serve as a basis for compound olfactory sensitivity. Thus, to account for differences in compound 152 perception, odor activity values are calculated by dividing the compound concentration by the olfactory threshold. This allows for more accurate comparisons of compound contributions to overall olfactory perception. Though differences in detection thresholds in water, air, or other substrates and variations between studies and consumers make comparisons less precise, general trends can be drawn. Reported detection thresholds of 1,8 cineole, eugenol, linalool, and methyl chavicol in water are 1.1, 0.71, 0.087, and 6.0 µg·L‒1, respectively (Czerny et al., 2008), while recognition thresholds are 4.6, 2.5, 0.17, and 16 µg·L‒1, respectively (Czerny et al., 2008; Zeller and Rychlik, 2006), although linalool recognition threshold has also been reported as 5.0 µg·L‒1 (Zeller and Rychlik, 2006). Therefore, in water, a higher methyl chavicol concentration is needed to be perceived as equally as 1,8 cineole, eugenol, or linalool. Both the detection and recognition thresholds for linalool are generally lower than 1,8 cineole, eugenol, and methyl chavicol, therefore, less linalool is needed to be perceived equally. In our experiment, linalool concentrations were lower than the other compounds measured and methyl chavicol concentrations where higher, therefore, both compounds still had significant contributions to overall basil aroma (Fig. IV-2). In addition to variation in compound perceptibility, likeability of compounds is an additional factor to consider in sensory analysis. To investigate the aroma acceptance of basil VOCs, researchers trained panelists to recognize linalool, 1,8 cineole, and eugenol by smelling progressively increasing concentrations of the standards (D’Antuono et al., 2007). Panelists then smelled diluted essential oil extracts from 24 basil cultivars and evaluated the perceived intensity of the compounds and overall acceptance. Though aroma acceptance varied greatly across cultivars, in general, high acceptance was not related to 1,8 cineole concentrations but correlated with high concentrations of linalool and low concentrations of eugenol. However, there were 153 exceptions to this correlation, leading the authors to conclude that a balanced volatile concentration is needed for the greatest consumer acceptance (D’Antuono et al., 2007). Though the ratios between compounds in our study was not as varied as the D’Antuono et al. (2007) study due to the use of one cultivar, ‘Nufar’, differing ratios of compounds did occur. In our study, the ratio (%) of 1,8 cineole:eugenol:linalool:methyl chavicol concentration for plants grown under 100 µmol·m‒2·s‒1 was 18:21:3:58 while plants growing under 600 µmol·m‒2·s‒1 had a lower proportion of methyl chavicol with ratios of 21:30:9:40. However, the total concentration of 1,8 cineole, eugenol, linalool, and methyl chavicol was nearly 3-fold higher in plants grown under 600 µmol·m‒2·s‒1 compared to those under 100 µmol·m‒2·s‒1 (Fig. IV-2). In our study, this total increase probably played a larger role in consumer preference than the ratio of terpenoids and phenylpropanoids. Consumer preferences. Consumer perception is influenced by many sensory modes including touch, sight, taste, and smell, where both taste and smell contribute to overall flavor. Though the exact number is unknown, researchers have estimated that consumers can generally distinguish 5,000 to 30,000 different odor qualities by smell, making their olfactory sense more diverse than the five basic tastes (Choi and Han, 2015). Although consumers can distinguish between many odors, they are more sensitive to flavor concentrations (Choi and Han, 2015). This was apparent in our data, as the aroma of basil grown under 100 µmol·m‒2·s‒1 was the least-preferred while aroma likability was similar among plants grown under 200 to 600 µmol·m‒2·s‒1 (Fig. IV-4). We postulate the reduced aroma preference of plants grown under 100 µmol·m‒2·s‒1 was due to the low VOC concentration, thus, weak aroma. While flavor and aftertaste preferences were greatest in plants grown under 200 µmol·m‒2·s‒1, as VOC concentration increased with increasing radiation 154 intensity, the flavor and aftertaste of plants grown under 600 µmol·m‒2·s‒1 were not as well liked, suggesting a greater sensitivity to high VOC concentration contributions to flavor compared to aroma (Figs. IV-2 and IV-4). Bitterness is generally negatively correlated with consumer preference, and is associated with harmful substances (Reineccius, 2005; Choi and Han, 2015). Therefore, the increased bitterness reported in basil grown under 400 and 600 µmol·m‒2·s‒1 may have contributed to the flavor and aftertaste preferences (Fig. IV-4). Additionally, attributes such as texture and color contribute to overall consumer preference. Basil grown under 600 µmol·m‒2·s‒1 exhibited the least-liked color and had among the lowest appearance and texture likeability (Figs. IV-5 and IV-6). The color was described by panelists as “brown” (Figs. IV-5 and IV-7) with symptoms of leaf damage due to the high radiation intensity. Additionally, consumers described the texture as “chewy” and “wilted”. This may be due to greater stomatal opening and gas exchange of plants grown under higher irradiances, which could have increased water loss and desiccation in the time between harvest and panelist evaluation (Davies and Kozlowski, 1975). Comparing compound concentrations and preferences. In addition to component 1 of the principal component analysis separating basil grown under 600 from 200 µmol·m‒2·s‒1, basil grown under 600 µmol·m‒2·s‒1 was correlated with 1,8 cineole, eugenol, and linalool concentrations (Fig. IV-8). Indeed, our GCMS analysis determined that basil grown under 600 µmol·m‒2·s‒1 had the highest 1,8 cineole, eugenol, and linalool concentrations (Fig. IV-2). However, contrary to our hypothesis that consumers would prefer basil with higher VOC concentrations, these higher concentrations were negatively correlated with sensory preference characteristics including bitterness/sweetness and color (Fig. IV-9). 155 Basil grown under 200 µmol·m‒2·s‒1 was correlated with consumer preferences for aftertaste, bitterness/sweetness, color, flavor, overall liking, and texture (Fig. IV-8). This corresponds to the consumer sensory preference values discussed previously, where 200 µmol·m‒2·s‒1 grown basil had among the highest Likert preference values for aftertaste, bitterness/sweetness, color, flavor, overall liking, and texture preference (Figs. IV-3, IV-4, and IV-6). Conclusions Overall, increasing radiation intensity during production increased terpenoid (1,8 cineole and linalool) and phenylpropanoid (eugenol) concentrations while CO2 concentration had no effect. Contrary to our hypothesis, increased VOC concentrations were not correlated with consumer preference. However, overall liking was correlated with aftertaste and flavor. However, increasing VOC concentrations to increase flavor did not improve flavor preference. Many consumer sensory preference characteristics (aftertaste, bitterness/sweetness, color, flavor, overall liking, and texture) were correlated with basil grown under a radiation intensity of 200 µmol·m‒2·s‒1, which had among the highest consumer preference values. This leads us to conclude that consumers prefer to detect the characteristic basil flavor made up of 1,8 cineole, eugenol, and linalool, that is not as prevalent in basil grown under 100 µmol·m‒2·s‒1, but too high of VOC concentrations when grown under 400 and 600 µmol·m‒2·s‒1 lead to a lower consumer preference. 156 APPENDICES 157 APPENDIX A TABLES AND FIGURES 158 Table IV-1. Target radiation intensity, actual radiation intensity, and average daily air, canopy, and substrate temperatures (mean ± SD) during the seedling growth stage (2 weeks). Chamber 1 2 1 2 Rep 1 Rep 2 z Data not collected Radiation intensity (µmol·m‒2·s‒1) Actual 102 ± 0 191 ± 0 429 ± 1 577 ± 2 94 ± 4 188 ± 2 432 ± 1 615 ± 2 99 ± 2 184 ± 3 384 ± 6 555 ± 9 88 ± 2 191 ± 3 394 ± 7 589 ± 11 Target 100 200 400 600 100 200 400 600 100 200 400 600 100 200 400 600 Air 23.0 ± 0.0 22.9 ± 1.8 23.0 ± 0.4 23.0 ± 1.4 Temperature (°C) - z Canopy Substrate 24.4 ± 0.9 21.9 ± 0.7 26.4 ± 1.0 27.4 ± 1.7 24.1 ± 1.3 28.7 ± 1.7 25.2 ± 1.9 24.5 ± 0.9 22.2 ± 1.0 24.0 ± 1.1 22.0 ± 0.9 26.9 ± 1.5 23.7 ± 1.4 27.7 ± 2.1 24.6 ± 1.9 24.2 ± 0.5 21.7 ± 0.6 26.5 ± 0.9 27.4 ± 1.6 24.3 ± 1.5 29.3 ± 2.0 25.3 ± 1.9 24.8 ± 0.9 22.5 ± 0.8 24.1 ± 0.8 22.7 ± 1.0 26.9 ± 1.5 23.5 ± 1.4 27.9 ± 2.2 24.6 ± 1.7 - 159 ) m n · 1 - s · 2 - m · l o m µ ( y t i s n e d x u l f n o t o h p e v i t a l e R 4 3 2 1 0 452 nm 603 nm 662 nm 100 µmol·m-2·s-1 200 µmol·m-2·s-1 400 µmol·m-2·s-1 600 µmol·m-2·s-1 400 500 600 700 800 Wavelength (nm) Figure IV-1. Spectral quality of light-emitting diode (LED) fixtures providing 20:40:40 blue:green:red radiation ratios (%), a red:far-red ratio of 13:1, and target radiation intensities of 100, 200, 400, or 600 µmol·m‒2·s‒1. 160 n o i t a r t n e c n o c e l o e n C 8 i , 1 ) M D 1 - g m · g n ( n o i t a r t n e c n o c l o o l a n L i ) M D 1 - g m · g n ( n o i t a r t n e c n o c l o n e g u E ) M D 1 - g m · g n ( n o i t a r t n e c n o c l o c i v a h c l y h t e M ) M D 1 - g m · g n ( 2000 1800 1600 1400 1200 1000 800 600 400 200 800 600 400 200 0 3000 2000 1000 0 4000 3000 2000 1000 0 0 Figure IV-2. Concentrations [ng·mg‒1 dry mass (DM)] of 1,8 cineole (A), linalool (B), eugenol 161 y = 2.16x + 1.20E2 r2 = 0.337*** y = 1.21x - 1.10E2 r2 = 0.341*** y = 1.42E-2x2 + -6.58x + 1.00E3 R2 = 0.179*** A B C D 200 600 Radiation intensity (µmol·m-2·s-1) 400 Figure IV-2 (cont’d). (C), and methyl chavicol (D) of sweet basil ‘Nufar’ (Ocimum basilicum) seedlings grown under 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD) for two weeks. Each symbol represents the mean of 20 plants ± SE. Lines represent linear or quadratic regression. *** indicates significant at P ≤ 0.001. 162 g n i t a r t r e k L n a e M i 7.0 6.8 6.6 6.4 6.2 6.0 5.8 5.6 1.0 0 Overall liking a ab ab b 600 200 400 Radiation intensity (µmol·m-2·s-1) Figure IV-3. Mean overall liking of sweet basil ‘Nufar’ (Ocimum basilicum) grown under 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD), based on a 9- point Likert scale ranging from dislike extremely (1) to like extremely (9). Means not followed by the same letter are significantly different by Tukey’s honestly significant difference test (P< 0.05). Each symbol represents 90 responses ± SD. 163 g n i t a r t r e k L n a e M i g n i t a r t r e k i L n a e M g n i t a r t r e k i L n a e M g n i t a r t r e k L n a e M i 7.2 7.0 6.8 6.6 6.4 6.2 6.0 5.8 5.6 1.0 5.0 4.8 4.6 4.4 4.2 4.0 3.8 1.0 7.0 6.8 6.6 6.4 6.2 6.0 5.8 5.6 1.0 4.6 4.4 4.2 4.0 3.8 3.6 3.4 3.2 1.0 0 Figure IV-4. Mean flavor (A), aftertaste (B), and aroma (C) of sweet basil ‘Nufar’ (Ocimum 164 ab ab b ab a Flavor ab a Aftertaste b A B b b a Aroma a C ab Bitterness/Sweetness a D bc c 200 600 Radiation intensity (µmol·m-2·s-1) 400 Figure IV-4 (cont’d). basilicum) grown under 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD), based on a 9-point Likert scale ranging from dislike extremely (1) to like extremely (9) and the bitterness/sweetness (D) based on a 9-point Likert scale ranging from extremely bitter (1) to extremely sweet (9). Means not followed by the same letter are significantly different by Tukey’s honestly significant difference test (P< 0.05). Each symbol represents 90 responses ± SD. 165 Figure IV-5. Leaves of basil used in the consumer sensory analysis panel. Sweet basil ‘Nufar’ (Ocimum basilicum) was grown under radiation intensities of 100, 200, 400, or 600 µmol·m– 2·s–1 for a 16-h photoperiod to create daily light integrals of 6, 12, 23, or 35 mol·m‒2·d‒1 for two weeks after sowing. 166 g n i t a r t r e k i l n a e M g n i t a r t r e k i l n a e M g n i t a r t r e k i l n a e M g n i t a r t r e k i l n a e M 7.4 7.2 7.0 6.8 6.6 6.4 6.2 6.0 1.0 7.6 7.4 7.2 7.0 6.8 6.6 6.4 6.2 1.0 7.6 7.2 6.8 6.4 6.0 5.6 5.2 4.8 1.0 7.2 7.0 6.8 6.6 6.4 6.2 6.0 1.0 0 Figure IV-6. The mean appearance (A), color (B), leaf size (C), and texture (D) of sweet basil 167 Appearance a ab a b a Color a Leaf size a Texture ab b a c ab A B C D ab b a b 200 600 Radiation intensity (µmol·m-2·s-1) 400 Figure IV-6 (cont’d). ‘Nufar’ (Ocimum basilicum) grown under 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD), based on a 9-point Likert scale ranging from dislike extremely (1) to like extremely (9). Means not followed by the same letter are significantly different by Tukey’s honestly significant difference test (P< 0.05). Each symbol represents 90 responses ± SD. 168 0 0 1 0 0 2 0 0 4 Frequency 0 5 10 15 20 25 30 35 40 45 50 0 6 0 Chi2 2 10.7 13.6 15.3 9.2 15.3 11.3 45.8 21.3 8.4 8.4 8.4 8.8 8.3 12.4 9.3 9.3 9.3 8.9 Bitter Leaves Buy Spice Enjoyable Subtle Small Tiny Mouth Reminds Quickly Yellow Vibrant Brown Wilted Spicier Odd Chewy Figure IV-7. Frequency heat map of words used to describe sweet basil ‘Nufar’ (Ocimum basilicum) grown under radiation intensities of 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD). Chi2 describes the goodness of fit, phylogenetic trees depict the cluster analysis relationship between words used and the relationship between words used and samples. All words reported exhibited differences P< 0.05 based on 90 respondents. 169 Figure IV-8. Principal component analysis (PCA) showing the biplot differentiation of sweet basil ‘Nufar’ (Ocimum basilicum) grown under 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD), based on consumer sensory preferences (n = 90) and the concentration of two terpenoids and two phenylpropanoids (n = 20). 170 1,8 Cineole Linalool Eugenol Methyl chavicol Overall liking Appearance Aroma Flavor Texture Color Leaf size Bitterness/Sweetness Aftertaste l o c i v a h c l y h t e M g n i k i l l l a r e v O e c n a r a e p p A l o o l a n i L l o n e g u E a m o r A r o v a l F e r u t x e T e l o e n i C 8 1 , * * - * - ** * ** - * * * * - - * ** - - * - - s s e n t e e w S / s s e n r e t t i B e t s a t r e t f A e z i s f a e L r o l o C * * * ** * - - * - * - Correlation -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 Figure IV-9. Heat map illustrating the correlation between volatile organic compounds (VOCs) and sensory preference characteristics of sweet basil ‘Nufar’ (Ocimum basilicum) grown under radiation intensities of 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD). Blue and red represent negative and positive correlations, respectively. Asterisks indicate significant correlations based on Pearson’s correlation *, P< 0.05; **, P<0.01. 171 APPENDIX B CONSUMER SENSORY EVALUATION SURVEY 172 Part 1. Basil Sensory Questions: Note: These questions were be repeated for each sample. Please taste the basil sample and answer the following questions: How do you like the sample OVERALL? Dislike extremely Dislike moderately Dislike very much Dislike slightly Neither like nor dislike Like slightly Like moderately Like very much Like extremely What do you LIKE about this sample? (open ended) What do you DISLIKE about this sample? (open ended) For each question below, please mark the box which best describes your opinion of the basil sample: Dislike extremely Dislike very much Dislike moderately Dislike slightly Like slightly Like moderately Overall appearance Overall aroma Overall flavor Overall texture Color Leaf size Please rate the level of bitterness/sweetness of the basil sample: Extremely bitter Moderately bitter Slightly bitter Neutral Slightly sweet 173 Like very much Like extremely Neither like nor dislike Moderately sweet Very sweet Please describe your experience of aftertaste in the basil sample: Dislike very much Dislike moderately None detected Dislike slightly Like slightly Like moderately Like very much Please describe the taste and share any additional comments about this basil sample (feel free to compare this sample to other samples). Part 2. Basil Sensory Survey: Q1 In the past four months (since June, 2018) have you eaten any of the following culinary herbs? Check all that you have eaten. • Fresh parsley • Dried parsley • Fresh basil • Dried basil • Fresh cilantro • Dried cilantro • Fresh thyme • Dried thyme Q2 How often have you eaten fresh basil in the past four months? (This includes fresh basil used in a variety of dishes) More than once a week • Once or twice a week • Once or twice a month • Once every other month or less Q3 In the past four months, have you grown your own basil? • Yes • No Q4 If you grew your own basil, did you grow it from: (check all that apply) • Seed • Small transplant • Potted plant • I did not grow my own basil 174 Q5 In the past four months, how often have you purchased basil? • Never • Weekly or almost weekly • Once or twice a month • Once or twice in the past four months Q6 Did you buy (check all that you have purchased since June, 2018)? • Cut basil (e.g. in a clamshell) • Potted basil • I did not buy fresh basil Q7 Where do you buy most of your fresh produce (fruits, vegetables, herbs)? Check just one. • National large grocery chain (e.g. Walmart or Kroger) • Independent grocery store • Roadside stand/at the farm • Grocery store specializing in one type of ethnic food • Subscription meal kit delivery (e.g. Blue Apron or HelloFresh) • CSA or community supported agriculture • Farmers market • Online order and pick up at store • Online order and deliver to home • Other, please specify Q8 Where do you buy most of your fresh herbs? Check just one. • National large grocery chain (e.g. Walmart or Kroger) • Independent grocery store • Roadside stand/at the farm • Grocery store specializing in one type of ethnic food • Subscription meal kit delivery (e.g. Blue Apron or HelloFresh) • CSA or community supported agriculture • Farmers market • Online order and pick up at store • Online order and deliver to home • Other, please specify 175 Q9 How important are the following characteristics when you buy culinary herbs? Drag the bar to the spot that corresponds to your opinion. Not at all important Slightly important 0 Extremely important 10 20 30 40 50 60 70 80 90 100 Moderately important important Very Large size Small size Aroma No visible blemishes or bruises Freshness Color Location grown Sweetness Flavor Texture Organically grown 176 Q10 To what extent do you agree or disagree with the following statements? Disagree Somewhat Somewhat agree Agree Strongly disagree disagree Neither agree nor disagree Strongly agree I like to serve herbs at or in many meals. Meals aren't enjoyable without herbs. I think herbs are important in my diet. I know a lot about herbs. I usually purchase the most expensive herbs. Local herbs taste better those coming from other regions. US grown herbs taste better than imported herbs. I prefer to buy the same herbs regardless of other choices. At the place of purchase, I inspect the herbs for defects. 177 Please check "somewhat agree" to insure each person is reading every question. At the place of purchase, I check the leaf color of the herbs. I can recognize many types of herbs. Q14 In what year were you born? ▼ 1920 (1) ... 2000 (81) Q15 What is your gender? • Male • Female • Other/prefer not to answer Q16 Not counting yourself, how many other adults (age 19 years and older) live in your household? ▼ 0 (1) ... 10 (11) Q17 How many children (age 18 years and under) live in your household? ▼ 0 (1) ... 10 (11) Q18 What is your ethnicity (ethnic heritage)? Please select all that apply. • White/Caucasian • African American • Hispanic • Asian • Native American • Pacific Islander • Other _______ 178 Q19 What is the highest level of education you have completed (please choose one)? • Less than High School • High School/GED • Some College • 2-year College Degree • 4-year College Degree • Master’s Degree • Doctoral Degree • Professional Degree (JD, MD) Q20 What was your approximate total family or household gross income for 2017 (please choose one)? 179 • Less than $19,999 • $20,000 to $39,999 • $40,000 to $59,999 • $60,000 to $79,999 • $80,000 to $99,999 • $100,000 to $119,999 • $120,000 to $139,999 • $140,000 to $159,999 • $160,000 to $179,999 • $180,000 to $199,999 • $200,000 or more • Prefer not to answer APPENDIX C CONSUMER SENSORY EVALUATION INTERNAL REVIEW BOARD APPROVAL 180 181 182 183 LITERATURE CITED 184 LITERATURE CITED Arena, E., N. Guarrera, S. Campisi, and C.N. Asmundo. 2006. Comparison of odour active compounds detected by gas-chromatography–olfactometry between hand-squeezed juices from different orange varieties. Food Chemistry, 98(1):59–63. Anterola, A.M. and N.G. Lewis. 2002. Trends in lignin modification: A comprehensive analysis of the effects of genetic manipulations/mutations on lignification and vascular integrity. Phytochemistry, 61(3):221–294. Bernhardt, B., L. Sipos, Z. Kókai, A. Gere, K. Szabó, J. Bernáth, and S.Z. Sárosi. 2015. Comparison of different Ocimum basilicum L. gene bank accessions analyzed by GC–MS and sensory profile. Industrial Crops and Products, 67:498–508. Buttery, R. G., D.R. Black, D.G. Guadagni, L.C. Ling, G. Connolly, and R. Teranishi. 1974. California bay oil. I. Constituents, odor properties. Journal of Agricultural and Food Chemistry, 22(5):773–777. Calín-Sánchez, Á., K. Lech, A. Szumny, A. Figiel, and Á.A. Carbonell-Barrachina. 2012. Volatile composition of sweet basil essential oil (Ocimum basilicum L.) as affected by drying method. Food Research International, 48(1):217–225. Chang, X., P.G. Alderson, and C.J. Wright. 2008. Solar irradiance level alters the growth of basil (Ocimum basilicum L.) and its content of volatile oils. Environmental and Experimental Botany, 63(1−3):216−223. Chang, X., P.G. Alderson, T.A. Hollowood, L. Hewson, and C.J. Wright. 2007. Flavour and aroma of fresh basil are affected by temperature. Journal of the Science of Food and Agriculture, 87(7):1381–1385. Choi, N.E. and J.H. Han. 2015. How flavor works: The science of taste and aroma. John Wiley & Sons, Chichester, UK. Christie, P.J., M.R. Alfenito, and V. Walbot. 1994. Impact of low-temperature stress on general phenylpropanoid and anthocyanin pathways: Enhancement of transcript abundance and anthocyanin pigmentation in maize seedlings. Planta, 194(4):541–549. Czerny, M., M. Christlbauer, M. Christlbauer, A. Fischer, M. Granvogl, M. Hammer, C. Hartl, N.M. Hernandez, and P. Schieberle. 2008. Re-investigation on odour thresholds of key food aroma compounds and development of an aroma language based on odour qualities of defined aqueous odorant solutions. European Food Research and Technology, 228 (2):265–273. D'Antuono, L.F., S. Elementi, R. Neri, and D. Sciannimanica. 2007. Sensory characterization of sweet basil essential oil. Acta Horticulturae. 741:109–113. 185 Davies, W.J. and T.T. Kozlowski. 1975. Stomatal responses to changes in light intensity as influenced by plant water stress. Forest Science, 21(2):129–133. Deschamps, C. and J.E. Simon. 2006. Terpenoid essential oil metabolism in basil (Ocimum basilicum L.) following elicitation. Journal of Essential Oil Research, 18(6):618–621. De Vincenzi, M., M. Silano, A. De Vincenzi, F. Maialetti, and B. Scazzocchio. 2002. Constituents of aromatic plants: Eucalyptol. Fitoterapia, 73(3):269–275. Díaz‐Maroto, M.C., E. Sánchez Palomo, L. Castro, M.A. González Viñas, and M.S. Pérez‐ Coello. 2004. Changes produced in the aroma compounds and structural integrity of basil (Ocimum basilicum L.) during drying. Journal of the Science of Food and Agriculture, 84(15):2070–2076. Dou, H., G. Niu, M. Gu, and J.G. Masabni. 2018. Responses of sweet basil to different daily light integrals in photosynthesis, morphology, yield, and nutritional quality. HortScience, 53(4):496–503. Dris, R. and S.M. Jain. 2004. Production practices and quality assessment of food crops. Kluwer Academic, Dordrecht, Netherlands. Eaves, J. and S. Eaves. 2018. Comparing the profitability of a greenhouse to a vertical farm in Quebec. Canadian Journal of Agricultural Economics/Revue Canadienne D'Agroeconomie, 66(1):43–54. Enfissi, E.M.A., P.D. Fraser, L.M. Lois, A. Boronat, W. Schuch, and P.M. Bramley. 2005. Metabolic engineering of the mevalonate and non-mevalonate isopentenyl diphosphate- forming pathways for the production of health-promoting isoprenoids in tomato. Plant Biotechnology Journal, 3(1):17–27. Estévez, J.M., A. Cantero, C. Romero, H. Kawaide, L.F. Jiménez, T. Kuzuyama, H. Seto, Y. Kamiya, and P. León. 2000. Analysis of the expression of CLA1, a gene that encodes the 1-deoxyxylulose 5-phosphate synthase of the 2-C-methyl-D-erythritol-4-phosphate pathway in Arabidopsis. Plant Physiology, 124(1):95–103. Falahi, H., M. Sharifi, H.Z. Maivan, and N.A. Chashmi. 2018. Phenylethanoid glycosides accumulation in roots of Scrophularia striata as a response to water stress. Environmental and Experimental Botany, 147:13–21. Fritz, C., N. Palacios‐Rojas, R. Feil, and M. Stitt. 2006. Regulation of secondary metabolism by the carbon–nitrogen status in tobacco: Nitrate inhibits large sectors of phenylpropanoid metabolism. The Plant Journal, 46(4):533–548. Gang, D.R., J. Wang, N. Dudareva, K.H. Nam, J.E. Simon, E. Lewinsohn, and E. Pichersky. 2001. An investigation of the storage and biosynthesis of phenylpropenes in sweet basil. Plant Physiology, 125(2):539–555. 186 Gomez, C., C.J. Currey, R.W. Dickson, H. Kim, R. Hernandez, N.C. Sabeh, R.E. Raudales, R.G. Brumfield, A. Laury-Shaw, A.K. Wilke, R.G. Lopez, and S.E. Burnett. 2019. Controlled environment food production for urban agriculture. HortScience, 54(9):1448–1458. Gong, Y.F., Z.H. Liao, B.H. Guo, X.F. Sun, and K.X. Tang. 2006. Molecular cloning and expression profile analysis of Ginkgo biloba DXS gene encoding 1-deoxy-D-xylulose 5- phosphate synthase, the first committed enzyme of the 2-C-methyl-D-erythritol 4- phosphate pathway. Planta Medica, 72(4):329–335. Goodman, W. and J. Minner. 2019. Will the urban agricultural revolution be vertical and soilless? A case study of controlled environment agriculture in New York City. Land Use Policy, 83:160–173. Haard, N.F. and B. Wasserman. 1976. Induction of phenylalanine ammonia lyase in sweet potato (Ipomoea batatas) root by light irradiation and black rot infection. Physiological Plant Pathology, 8(2):207–213. Hälvä, S., L.E. Craker, J.E. Simon, and D.J. Charles. 1992. Light levels, growth, and essential oil in dill (Anethum graveolens L.). J. Herbs, Spices, Medicinal Plants, 1(1-2):47−58. Heo, J.W., D.H. Kang, H.S. Bang, S.G. Hong, C.H. Chun, and K.K. Kang. 2012. Early growth, pigmentation, protein content, and phenylalanine ammonia-lyase activity of red curled lettuces grown under different lighting conditions. Korean Journal for Horticultural Science, 30(1):6–12. Hoops, G.P., H.P. Olshansky, J. Tavares, and R. Rosen. 2018. Copenhagen's case for urban farming: A feasibility study. Bachelor of Science thesis, Worcester Polytechnic Institute. Iijima, Y., R. Davidovich-Rikanati, E. Fridman, D.R. Gang, E. Bar, E. Lewinsohn, and E. Pichersky. 2004. The biochemical and molecular basis for the divergent patterns in the biosynthesis of terpenes and phenylpropenes in the peltate glands of three cultivars of basil. Plant Physiology, 136(3):3724–3736. Juliani, H. R., J.E. Simon, M.R. Ramboatiana, O. Behra, A. Garvey, and I. Raskin. 2002. Malagasy aromatic plants: Essential oils, antioxidant and antimicrobial activities. Acta Horticulturae 629:77–81. Karapinar, M. and S. Aktug. 1987. Inhibition of foodborne pathogens by thymol, eugenol, menthol and anethole. International Journal of Food Microbiology, 4(2):161–166. Kim, B.R., S.U. Kim, and Y.J. Chang. 2005. Differential expression of three 1-deoxy-D- xylulose-5-phosphate synthase genes in rice. Biotechnology Letters, 27(14):997–1001. Kuhn, D.N., J. Chappell, A. Boudet, and K. Hahlbrock. 1984. Induction of phenylalanine ammonia-lyase and 4-coumarate: CoA ligase mRNAs in cultured plant cells by UV light or fungal elicitor. Proceedings of the National Academy of Sciences, 81(4):1102–1106. 187 Kumar, R., S. Sharma, and V. Pathania. 2013. Effect of shading and plant density on growth, yield and oil composition of clary sage (Salvia sclarea L.) in north western Himalaya. Journal of Essential Oil Research, 25(1):23−32. Leyva, A., J.A. Jarillo, J. Salinas, and J.M. Martinez-Zapater. 1995. Low temperature induces the accumulation of phenylalanine ammonia-lyase and chalcone synthase mRNAs of Arabidopsis thaliana in a light-dependent manner. Plant Physiology, 108(1):39–46. Li, Y.L., L.E. Craker, and T. Potter. 1995. Effect of light level on essential oil production of sage (Salvia officinalis) and thyme (Thymus vulgaris). Acta Horticulturae, 426:419−426. Metz, B., O.R. Davidson, P.R. Bosch, R. Dave, and L.A. Meyer. 2007. Climate change 2007: Mitigation of climate change. Cambridge University Press, Cambridge, UK and New York, NY. Moleyar V. and P. Narasimham. 1992. Antibacterial activity of essential oil components. International Journal of Food Microbiology, 16(4):337–342. Morales, M.R. and J.E. Simon. 1997. 'Sweet Dani': A new culinary and ornamental lemon basil. HortScience, 32(1):148–149. Muñoz-Bertomeu, J., I. Arrillaga, R. Ros, and J. Segura. 2006. Up-regulation of 1-deoxy-D- xylulose-5-phosphate synthase enhances production of essential oils in transgenic spike lavender. Plant Physiology, 142(3):890–900. Olsen, K.M., U.S. Lea, R. Slimestad, M. Verheul, and C. Lillo. 2008. Differential expression of four Arabidopsis PAL genes; PAL1 and PAL2 have functional specialization in abiotic environmental-triggered flavonoid synthesis. Journal of Plant Physiology, 165(14):1491– 1499. Patton, S. and D.V. Josephson. 1957. A method for determining significance of volatile flavor compounds in foods. Journal of Food Science, 22(3):316–318. Reineccius, G. 2005. Flavor chemistry and technology. CRC press, Boca Raton, FL. Reynolds, K. and N. Cohen. 2016. Beyond the kale: Urban agriculture and social justice activism in New York City. The University of Georgia Press, Athens, GA. Riov, J., S.P. Monselise, and R.S. Kahan. 1969. Ethylene-controlled induction of phenylalanine ammonia-lyase in citrus fruit peel. Plant Physiology, 44(5):631–635. Rivero, R.M., J.M. Ruiz, P.C. Garcia, L.R. Lopez-Lefebre, E. Sánchez, and L. Romero. 2001. Resistance to cold and heat stress: Accumulation of phenolic compounds in tomato and watermelon plants. Plant Science, 160(2):315–321. Rodríguez-Concepción, M. and A. Boronat. 2015. Breaking new ground in the regulation of the early steps of plant isoprenoid biosynthesis. Current Opinion in Plant Biology, 25:17–22. 188 Sangwan, N., B. Verman, K. Verma, and K. Dhindsa. 1990. Nematicidal activity of some essential plant oils. Pesticide Science, 28(3):331–335. Santos, A., G. Chierice, K. Alexander, A. Riga, and E. Matthews. 2009. Characterization of the raw essential oil eugenol extracted from Syzygium aromaticum L. Journal of Thermal Analysis and Calorimetry, 96(3):821–825. Sanyé-Mengual, E., F. Orsini, and G. Gianquinto. 2018. Revisiting the sustainability concept of urban food production from a stakeholders’ perspective. Sustainability, 10(7):2175. Schilmiller, A., F. Shi, J. Kim, A.L. Charbonneau, D. Holmes, A.D. Jones, and R.L. Last. 2010. Mass spectrometry screening reveals widespread diversity in trichome specialized metabolites of tomato chromosomal substitution lines. The Plant Journal, 62(3):391–403. Shaaya, E., U. Ravid, N. Paster, B. Juven, U. Zisman, and V. Pissarev. 1991. Fumigant toxicity of essential oils against four major stored-product insects. Journal of Chemical Ecology, 17(3):499–504. Seely, A.K. 2017. Determining aroma differences among basil, parsley, and dill grown under varied supplemental light wavelengths using consumer sensory and flash gas chromatograph-electronic nose analyses. Master’s thesis, Iowa State University, ProQuest Dissertations Publishing. Simon, J.E., M.R. Morales, W.B. Phippen, R.F. Vieira, and Z. Hao. 1999. Basil: A source of aroma compounds and a popular culinary and ornamental herb. p. 499–505. In: J. Janick (ed.). Perspectives on new crops and new uses. ASHS Press. Arlington, VA. Simpson, K., L.F. Quiroz, M. Rodriguez-Concepción, and C.R. Stange. 2016. Differential contribution of the first two enzymes of the MEP pathway to the supply of metabolic precursors for carotenoid and chlorophyll biosynthesis in carrot (Daucus carota). Frontiers in Plant Science, 7:1344. Singh, P., R.M. Kalunke, and A.P. Giri. 2015. Towards comprehension of complex chemical evolution and diversification of terpene and phenylpropanoid pathways in Ocimum species. Royal Society of Chemistry, 5(129):106886–106904. Sisk C., H. Shorey, R. Gerber, and L. Gaston. 1996. Semiochemicals that disrupt foraging by the Argentine ant (Hymenoptera: Formicidae): Laboratory bioassays. Journal of Economic Entomology, 89(2):381–385. Tangpao, T., H.H. Chung, and S. Sommano. 2018. Aromatic profiles of essential oils from five commonly used Thai basils. Foods, 7(11):175. Walters, K.J. and C.J. Currey. 2015. Hydroponic greenhouse basil production: Comparing systems and cultivars. HortTechnology, 25(5):645−650. 189 Walters, K.J., B.K. Behe, C.J. Currey, and R.G. Lopez. 2020. Historical, current, and future perspectives for controlled environment hydroponic food crop production in the United States. HortScience, 55(6):758−767. Weber, C.L. and H.S. Matthews. 2008. Food-miles and the relative climate impacts of food choices in the United States. Environmental Science and Technology, 42(10):3508–3513. Wong, P.P., M. Zucker, and L.L. Creasy. 1974. Induction of phenylalanine ammonia-lyase in strawberry leaf disks: Action spectra and effects of wounding, sucrose, and light. Plant Physiology, 54(5):659–665. Wright, L.P., J.M. Rohwer, A. Ghirardo, A. Hammerbacher, M. Ortiz-Alcaide, B. Raguschke, J.P. Schnitzler, J. Gershenzon, and M.A. Phillips. 2014. Deoxyxylulose 5-phosphate synthase controls flux through the methylerythritol 4-phosphate pathway in Arabidopsis. Plant Physiology, 165(4):1488–1504. Zeidler, C., D. Schubert, and V. Vrakking. 2013. Feasibility study: Vertical farm EDEN, DLR Institute of Space Systems, German Aerospece Center. Zeller, A. and M. Rychlik. 2006. Character impact odorants of fennel fruits and fennel tea. Journal of Agricultural and Food Chemistry, 54(10):3686–3692. 190 CHAPTER 5 HYDROPONIC BASIL PRODUCTION: TEMPERATURE INFLUENCES THE PROFILE OF VOLATILE ORGANIC COMPOUNDS, BUT NOT OVERALL CONSUMER PREFERENCE 191 Hydroponic basil production: Temperature influences the profile of volatile organic compounds, but not overall consumer preference Kellie J. Walters and Roberto G. Lopez Acknowledgements We gratefully acknowledge Sean Tarr, Nate DuRussel, Cassandra Johnny, Bridget Behe, Randy Beaudry, Philip Engelgau, and Alex Renny for assistance, PL Lighting for HPS lamps, JR Peters for fertilizer, Grodan for substrate, Hydrofarm for hydroponic production systems, Ludvig Svensson for shade cloth, and the Michigan State University Research Technology Support Facility Mass Spectrometry and Metabolomics Core for GCMS access and method guidance. This work was supported by Michigan State University AgBioResearch (including Project GREEEN GR19-019), the USDA National Institute of Food and Agriculture Hatch project nos. MICL02472, MICL02589, and MICL02473, and USDA-NIFA Specialty Crop Research Initiative award no. 2018-51181-28377, and The Fred C. Gloeckner Foundation. The use of trade names in this publication does not imply endorsement by Michigan State University of products named nor criticism of similar ones not mentioned. 192 Additional Index Words: 1,8 cineole, estragole, eucalyptol, eugenol, linalool, mean daily temperature (MDT), methyl chavicol, Ocimum basilicum, phenylpropanoid, sensory panel, terpenoid Abstract. Altering the growing temperature during controlled environment production not only influences crop growth and development, but can also influence volatile organic compound (VOC) production and subsequently, sensory attributes of culinary herbs. Therefore, the objectives of this study were to 1) quantify the influence of air mean daily temperature (MDT) and daily light integral (DLI) on key sweet basil (Ocimum basilicum) phenylpropanoid and terpenoid concentrations, 2) determine if differences in sensory characteristics due to differing MDTs and DLIs influence consumer preference, and 3) identify the sweet basil attributes consumers prefer. Thus, two-week old sweet basil ‘Nufar’ seedlings were transplanted into deep- flow hydroponic systems in greenhouses with target MDTs of 23, 26, 29, 32, or 35 °C and DLIs of 7, 9, or 12 mol·m‒2·d‒1. After three weeks, the two most recently mature leaves were harvested for gas chromatography mass spectrometry (GCMS) analysis and only plants grown under the highest DLI were harvested for consumer sensory analysis. Panel evaluations were conducted through a sliding-door with samples served individually while panelists answered Likert scale and open-ended quality attribute and sensory questions. The DLI did not influence VOC concentrations. Increasing MDT from 23 to 36 °C during production increased 1,8 cineole, eugenol, and methyl chavicol concentrations linearly and did not affect linalool concentration. The increases in phenylpropanoid (eugenol and methyl chavicol) were greater than increases in terpenoid (1,8 cineole) concentrations. However, these increases did not impact overall consumer 193 or flavor preference. The MDT during basil production did influence appearance, texture, and color preference of panelists. Taken together, MDT during production influenced both VOC concentrations and textural and visual attribute preference of basil but it did not influence overall consumer preference. Therefore, changing the MDT during production can be used to alter plant growth and development without significantly affecting consumer preference. Introduction Interest in controlled environment agriculture (CEA) is increasing due to demand for year-round and consistent supply of locally grown food, recent supply chain issues, food borne illnesses, and technological innovations improving economic feasibility of CEA (Busch et al., 2020; Gomez et al., 2019; Kolodinsky et al., 2020). Producing crops in controlled environments has the potential to improve production efficiencies by precisely manipulating the growing environment to obtain desired yield and quality attributes (Gomez et al., 2019). While technology exists to precisely control the greenhouse environment, knowledge of specific crop requirements that would allow growers to utilize the technology to its greatest potential is lacking. If known, increased production efficiencies and higher-quality products could be realized. Additionally, with potentially greater capital and operating costs for CEA compared to conventional (field) production, producing a higher-value product is highly advantageous. One potential method to increase the quality of a food crop is by improving flavor. In a recent survey, 90% of grower participants in the United States (U.S.) stated that their customers would pay more for crops with increased flavor (Walters et al., 2020). This allows us to hypothesize that consumers would prefer basil (Ocimum basilicum) with a more intense flavor. 194 The main contributors to basil flavor are volatile organic compounds (VOCs). Based on a preliminary analysis of ‘Nufar’ sweet basil secondary metabolite concentrations, two phenylpropanoids (eugenol and methyl chavicol) and two terpenoids (1,8 cineole and linalool) make up a large proportion of the total VOC content (data not shown). Eugenol is the major VOC in cloves (Syzygium aromaticum) and contributes a clove-like flavor and aroma to basil (Santos et al., 2009). The anise-like aroma and flavor characteristic of basil can be attributed to methyl chavicol (estragole; Simon et al., 1999). The monoterpenoid linalool is commonly found in many plants, contributing a flavor and aroma that has been described as having an aroma and flavor of floral or spicy (Arena, et al., 2006) or Fruit Loops® cereal. Finally, 1,8 cineole or eucalyptol has an aroma and flavor analogous to eucalyptus (Eucalyptus globulus), whose essential oil contains 70% to 80% 1,8 cineole (De Vincenzi et al., 2002). Increasing temperature can increase VOC production. For example, dill (Anethum graveolens) VOC content increased with increasing temperature (Hälvä et al., 1993). As temperature increases, enzymatic activity and the production of VOCs can increase to a point. Chang et al. (2005) determined that the concentration of VOCs in basil ‘Sweet Genovese’ grown at 25 or 30 °C for two weeks was three times higher than plants grown at 15 °C, but not all compounds responded the same. Basil grown at 15 °C had more camphor and trans-farnesene, while basil grown at 25 or 30 °C had higher eugenol and cis-ocimene, and temperature did not affect the concentration of 1,8-cineole or linalool. Another study reported that basil grown at 27 °C had a greater concentration of terpene VOCs compared to plants grown at 21 or 32 °C (Pogany et al., 1968). Growers often manipulate the growing environment to improve yield without considering how it will affect flavor, aroma, and ultimately consumer preference. Researchers have analyzed 195 changes in specific secondary metabolites in response to changing environmental conditions, but that may not necessarily correlate with consumer preference. Other factors influencing consumer preference include sensory thresholds of specific compounds, the interaction of these compounds, variation in sensory ability, and preference in general. Previous basil sensory evaluation studies have largely focused on aroma. Chang et al. (2007) conducted a study in which 21 trained panelists and 64 untrained consumer panelists both identified that basil plants grown at 25 °C had a stronger aroma than those grown at 15 °C. However, due to large variation in the frequency of basil use among consumer panelists, researchers were unable to draw conclusions on consumer aroma preference. An improved understanding of the extent individual environmental parameters, e.g., temperature or radiation intensity, have on consumer preference can aid growers in balancing consistently high-yielding, high-quality crops that customers enjoy consuming. This information could aid growers in minimizing quality differences throughout the year, potentially improving their ability to market their crop based on “improved flavor,” and finding a balance between crop quality and yield. Therefore, connecting the analytical data on the concentration of VOCs that contribute to flavor with consumer preferences is integral to improving crop flavor. Thus, the objectives of this study were to 1) quantify the influence of mean daily temperature (MDT) and daily light integral (DLI) on key basil phenylpropanoid and terpenoid concentrations, 2) determine if differences in sensory characteristics due to MDT and DLI are detectible by consumers, and 3) identify the sweet basil attributes consumers prefer. We hypothesized that increasing both MDT and DLI would increase VOC concentrations and impact sensory characteristics. Additionally, we hypothesized that consumers would prefer more flavorful basil. 196 Methods Growing environment. Sweet basil ‘Nufar’ seeds (Johnny’s Selected Seeds; Winslow, MA) were sown in stone wool cubes (2.5 × 2.5 × 4 cm, AO plug; Grodan, Roermond, Netherlands) and grown for two weeks with a target air temperature of 23 °C. The seeds and seedlings were irrigated daily and the growing environment was monitored, as reported in Chapter 3. Two weeks after sowing, seedlings were transplanted on 20 Oct., 2017 (rep 1) and 19 Sept. 2018 (rep 2) into systems and the growing environment was monitored and controlled as reported in Chapter 3. Five greenhouse compartments contained target air MDTs of 23, 26, 29, 32, or 35 °C. Each greenhouse contained three hydroponic systems under 0%, ~30%, or ~50% shade cloth (Solaro 3215 D O FB and Solaro 5220 D O; Ludvig Svensson, Kinna, Sweden) used to create target DLIs of 12, 9, or 7 mol·m‒2·d‒1, respectively, with actual DLIs and MDTs reported in Table V-1. GCMS analysis. Three weeks after transplant, the two most recent, fully mature leaves of five plants from each treatment were detached, frozen, and stored at -20 °C until gas chromatography mass spectrometry (GCMS) analysis as reported in Chapter 4 with a method derived from Schilmiller et al. (2010). Sensory analysis. The protocol for the sensory analysis portion of this study was approved by the Institutional Review Board of Michigan State University (MSU; East Lansing, MI, USA; STUDY00000169). The experimental procedure was thoroughly explained to all participants and a written informed consent was obtained from each prior to participation. Eighty-six participants 197 were recruited through the MSU Communication Arts and Sciences Paid Research Pool and screened prior to participation to ensure they had eaten basil in the four months prior. The sensory panel was conducted using the Sensory Evaluation Laboratory in the Department of Food Science and Human Nutrition at MSU. On 10 Oct. 2018 (rep 2), the two most recent, fully mature leaves of each plant were harvested 1 to 4 h prior to sampling to ensure freshness. The leaves were removed from the plant and rinsed in deionized water. The samples were dried through air-drying and gentle patting with a towel. One leaf was placed on a plate for panelist evaluation. The participants sampled the basil leaves as described in Chapter 4 and answered Likert, open-ended, and demographic questions (Chapter 4, Appendix B). Experimental design and statistical analysis. The experiment was organized in a split-plot design with each of five MDTs in separate greenhouse sections and three DLI treatments in each section. Only the highest DLI in each MDT treatment in replication 2 was used for consumer sensory analysis (Table V-1). The experiment was completed twice in time for GCMS analysis (reps 1 and 2) and once for consumer sensory analysis (rep 2). Analysis of variance was performed on VOC and Likert data using JMP (version 12.0.1, SAS Institute Inc., Cary, NC). When interactions were not present, data were pooled. Linear regression analyses were conducted on VOC data using Sigma Plot (version 11.0, Systat Software Inc., San Jose, CA) and Tukey’s honestly significant difference test (P < 0.05) was conducted on Likert data using JMP. Data was transformed to account for differences in values, variation, and sample size [(sample value – parameter average)/parameter SD], then principal component analysis was conducted (using JMP). Biplots were created combining principal component analysis with loading plots (JMP). Factor analysis was conducted to determine the 198 significant factors based on a rotated factor loading value of > 0.7. Correlations were determined by Pearson’s correlation coefficient P < 0.05 (JMP). Results VOC concentration. Due to the lack of a DLI×MDT interaction and main effect of DLI for any of the VOCs quantified, DLIs were pooled for GCMS analysis. As MDT increased from 23 to 36 °C, 1,8 cineole, eugenol, and methyl chavicol concentrations increased linearly; linalool concentration was not affected (Fig. V-1A-D). Eugenol and methyl chavicol concentration increased at similar rates as temperature increased [487 to 499 ng·mg‒1 dry mass (DM) per 1 °C], while 1,8 cineole concentration increased at a lower rate (40 ng·mg‒1 DM per 1 °C). Consumer preference. The mean age of consumer panelists was 32.4 years, with an average of 1.3 adults and 0.5 minors in the household. Fifty-eight percent of panelists were female, 41% were male, 67% were Caucasian, and 19% were Asian. Average household income was $50,000, and 64% had at least a 4-year post-secondary education. Consumers indicated no differences in flavor, aftertaste, aroma, leaf size, or overall preference among basil grown from 25 to 35 °C (data not shown). However, they preferred the appearance, texture, and color of basil grown at some of the higher MDTs compared to 25 °C (Figs. V-2A-C). Plants grown at a MDT of 25°C were considered less bitter (more sweet) than those grown at 31 °C (Fig. V-2B). Comparison – Biplot and correlations. 199 A principal component analysis comparison of VOCs and consumer sensory preferences, represented by a biplot including basil samples grown at 25, 27, 29, 31, and 35 °C, determined that two components accounted for 88.6% of the total data variability (Fig. V-3). Component 1 separated basil grown at low MDT (25 °C) from those grown at higher MDTs (e.g. 35 °C). Positive discriminating factors for component 1 were preference for appearance, texture, color, and leaf size while negative discriminating factors were bitterness/sweetness and aftertaste preference. Component 2 positive discriminating factors were flavor preference, overall liking, and 1,8 cineole, eugenol, and methyl chavicol concentration, while there was no significant negative discriminating factor. Appearance, texture, and color were positively correlated with each other, but appearance and texture were negatively correlated with bitterness/sweetness (increased sweetness; Fig. V-4). Aftertaste preference was positively correlated with bitterness/sweetness (increased sweetness). Additionally, methyl chavicol concentration was positively correlated with flavor preference and overall liking, and 1,8 cineole and eugenol concentrations were positively correlated with each other (Fig. V-4). Discussion Volatile organic compound concentration. The linear increase in 1,8 cineole, eugenol, and methyl chavicol concentrations in response to MDT (Fig. V-1) may be partially explained by a commonly referred to plant development temperature response curve. Temperature response curves are often used to describe the relationship between plant growth or development and MDT (Adams et al., 1997; Karlsson et al., 1988; Pramuk and Runkle, 2005; Tollenaar et al., 1979; Warner, 2020) but can 200 also be adapted to describe enzyme kinetics and other processes involving enzymatic regulation, including the biosynthesis of VOCs such as terpenoids and phenylpropanoids. Temperature plays a large role in plant enzymatic activity; as temperature increases, enzymatic activity increases up to an enzyme-dependent maximum. Above that point, many enzymes cease to function regularly (Cseke et al., 2016). Therefore, based on the temperature response curve for VOCs (Fig. V-1), MDTs in the present study fall in the linear range between Tb and Topt. The linear increase in 1,8 cineole concentration as MDT increased from 23 to 36 °C was less than the increases in eugenol and methyl chavicol concentrations. Similar results have been documented. Chang et al. (2005) reported that increasing the air temperature from 15 to 25 °C increased basil ‘Basil Sweet Genovese’ eugenol concentration while 1,8-cineole and linalool concentrations were unaffected. Though they documented no differences in 1,8 cineole concentrations, their temperature range was lower (15 to 25 °C) than the present study (23 to 36 °C). Differences in the magnitude of change (regression slope) between terpenoids and phenylpropanoids stem from the biosynthetic pathways. In addition to temperature affecting enzymatic catalysis rate, specific enzyme concentrations can be directly regulated by temperature. For example, phenylalanine ammonia lyase (PAL) catalyzes the first committed step of phenylpropanoid biosynthesis and is a key regulatory enzyme for many compounds in addition to phenylpropanoids. PAL catalyzes the removal of ammonia (NH3) from phenylalanine (Weisshar and Jenkins, 1998), and is positively regulated by decreasing temperatures (Leyva et al., 1995). However, this regulation affects many secondary metabolites and does not necessarily increase phenylpropanoids. For example, eugenol concentration in basil was higher when the growing temperature was higher (Chang et al., 2005). This may be caused by different PALs 201 playing a role in different biosynthetic pathways. However, terpene concentration is largely regulated by enzymatic activity and resource competition; the total amount of terpenes positively correlated with terpene synthase activities and negatively correlated with phenylpropanoid concentrations and PAL activity (Iijima et al., 2004). In Chapter 4, we found that VOC concentrations increased as radiation intensity increased from 100 to 600 µmol·m‒2·s‒1 PPFD (6 to 35 mol·m‒2·d‒1 DLI). The lack of differences between DLI treatments in the current study may be due to the more narrow DLI range (5.5 to 14.8 mol·m‒2·d‒1), increased radiation intensity variability due to solar radiation in greenhouse production compared to sole-source lighting, or a plant age of 5 weeks compared to 2 weeks in the previous study (Chapter 4). Consumer preferences and correlations. Though the concentrations of three compounds that contribute to the characteristic flavor of basil increased as MDT increased from 25 to 35 °C, consumer preference for flavor, aroma, and aftertaste did not change. Previously, researchers determined that consumers can detect an increase in basil aroma as growing temperatures increase from 15 to 25 °C, but there was no difference in consumer preference (Chang et al., 2007). However, consumers are more sensitive to flavor than aroma (Choi and Han, 2015). As we found in Chapter 4, though VOC concentrations increased as radiation intensity increased from 100 to 600 µmol·m‒2·s‒1 PPFD (6 to 35 mol·m‒2·d‒1 DLI), aroma of basil grown under 100 µmol·m‒2·s‒1 was the least-preferred and likability was similar among plants grown under 200, 400, or 600 µmol·m‒2·s‒1. Therefore, though consumers may perceive an increase in aroma intensity, if the aroma exhibits the characteristic aroma they expect, they may not have a concentration preference. 202 Previous research found that the concentration of linalool, 1,8 cineole, and eugenol of indoor-grown basil increased with increasing radiation intensity and the concentrations influenced consumer’s flavor preference (Chapter 4). However, in this study, while the concentrations of 1,8 cineole, eugenol, and methyl chavicol increased with increasing temperature, flavor preference was not influenced. Eugenol and methyl chavicol concentrations were higher in the greenhouse-grown plants in this study compared to the indoor-grown seedlings, with eugenol concentrations up to 225-fold higher and methyl chavicol concentrations ~1 to 6-fold higher. These differences may be due to plant age, as the seedlings grown indoors were two weeks old and the greenhouse-grown plants were five weeks old. Previous research demonstrated that terpenoid and phenylpropanoid concentrations of basil ‘Caesar’ were much greater in 4 to 6 week old plants compared to 2 to 3 week old plants (Carvalho et al., 2016). In contrast, 1,8 cineole concentrations overall were comparable between the seedling radiation intensity study and the current study, but the concentration increased more in response to increasing radiation intensity during seedling production than increasing MDT during finished production. Taken together, the higher phenylpropanoid concentrations and attenuated increase in 1,8 cineole concentration may account for the lack of differences in flavor preference attributed to MDT during production. Even though overall preference was not significantly different between basil grown at different MDTs, principal component and Pearson correlation analyses indicated that methyl chavicol concentration was positively correlated with flavor and overall preference (Fig. V-4). Thus, increasing the anise or licorice flavor correlated with improved consumer flavor and overall preference. Similarly, in a study comparing the oil aroma profile of basil, methyl chavicol was the compound most positively correlated with overall preference (Sheen et al., 1991). 203 Though bitterness has evolutionarily been an indicator of the presence of harmful secondary metabolites and used in plant defense mechanisms, it is a characteristic of many preferred foods and beverages today. However, it is still generally negatively correlated with consumer preference. In the present study, consumers indicated that basil grown at 31 °C was slightly bitter while basil grown at 25 °C was more neutral, neither sweet nor bitter (Fig. V-2D). Additionally, aftertaste preference was positively correlated with less bitterness (Fig. V-4). Researchers have determined that bitter tastes tend to be perceived after other tastes and linger longer contributing to aftertaste (Naim et al., 2002). Therefore, since increased bitterness is commonly associated with negative consumer preference, it can also be associated with a reduced aftertaste preference. Positive discriminating factors associated with higher MDTs are greater preference for appearance, texture, color, and leaf size (Fig. V-3). Though these were different among basil grown at different temperatures, these factors did not influence overall preference. While visual and textural attributes can contribute to overall consumer preference, unless the appearance, texture, or color is deemed unacceptable, the correlation data indicates that factors including flavor preference may be larger indicators of overall preference (Fig. V-4). Conclusions Increasing MDT from 25 to 35 °C during basil production increased 1,8 cineole, eugenol, and methyl chavicol concentrations linearly. This temperature range did not affect linalool concentration. Increases in phenylpropanoids (eugenol and methyl chavicol) were greater than increases in terpenoid (1,8 cineole) concentrations. However, these increases did not influence overall consumer preference or flavor preference. MDT during basil production did influence 204 appearance, texture, and color preference although these parameters, again, did not influence overall consumer preference. Taken together, MDT during production does influence both VOC concentrations and textural and visual attributes of basil but did not influence overall consumer preference. Therefore, changing MDT during production can be used to alter plant growth, morphology, development, and yield (Chapter 3), without influencing consumer preference. 205 APPENDIX 206 Table V-1. Mean daily light integral (DLI; mol·m‒2·d‒1 ± SD) and mean daily air temperature (MDT), leaf, and nutrient solution temperatures during the three-week growing period for sweet basil (Ocimum basilicum ‘Nufar’) with two replications in over time. Data were collected every 15 s with means logged every hour. DLI Target Rep. 1 2 z data not collected y partial data reported 12 9 7 12 9 7 12 9 7 12 9 7 12 9 7 12 9 7 12 9 7 12 9 7 12 9 7 12 9 7 Actual 12.9 ± 1.1 10.1 ± 1.0 7.4 ± 0.8 11.1 ± 1.1 9.2 ± 1.0 8.0 ± 0.8 11.3 ± 1.2 8.9 ± 1.1 6.6 ± 0.9 13.2 ± 1.3 9.2 ± 1.0 7.2 ± 0.7 12.0 ± 1.3 8.5 ± 0.9 5.5 ± 0.8 12.4 ± 3.2 9.3 ± 2.3 7.8 ± 2.0 13.4 ± 3.5 9.9 ± 2.9 7.1 ± 2.3 14.0 ± 4.1 9.3 ± 2.9 6.1 ± 2.1 14.8 ± 3.8 9.8 ± 2.5 7.3 ± 2.0 10.3 ± 2.7 7.7 ± 2.4 - - - - - - - - z - Temperature °C Leaf Target Air Actual Air Solution 23.0 ± 1.3 26.3 ± 1.3 21.7 ± 1.4 23.0 ± 1.3 21.5 ± 1.1 23.0 ± 1.3 21.4 ± 1.3 25.6 ± 0.6 27.7 ± 0.9 23.9 ± 0.7 25.6 ± 0.6 23.4 ± 0.7 25.6 ± 0.6 23.5 ± 0.7 28.9 ± 1.9 30.5 ± 0.8 26.0 ± 1.0 28.9 ± 1.9 25.9 ± 0.9 28.9 ± 1.9 25.7 ± 1.0 31.4 ± 0.7 35.7 ± 1.2 28.7 ± 0.8 27.3 ± 0.8 31.4 ± 0.7 31.4 ± 0.7 26.2 ± 0.8 35.7 ± 0.8 36.6 ± 0.8 30.3 ± 0.8 29.6 ± 0.7 35.7 ± 0.8 35.7 ± 0.8 29.2 ± 0.6 24.9 ± 2.2 26.5 ± 1.5 24.6 ± 1.7 22.9 ± 1.0 y 24.9 ± 2.2 24.9 ± 2.2 23.0 ± 1.2 27.0 ± 1.2 27.1 ± 2.0 26.4 ± 0.8 27.0 ± 1.2 25.3 ± 1.1 27.0 ± 1.2 24.9 ± 1.0 29.2 ± 0.6 30.2 ± 0.9 26.8 ± 0.6 29.2 ± 0.6 26.6 ± 0.8 29.2 ± 0.6 26.6 ± 0.6 31.3 ± 1.0 32.0 ± 1.8 28.9 ± 0.3 28.7 ± 0.5 31.3 ± 1.0 31.3 ± 1.0 28.6 ± 0.6 y 34.6 ± 1.8 33.6 ± 0.9 31.6 ± 1.3 30.2 ± 0.7 34.6 ± 1.8 34.6 ± 1.8 29.5 ± 0.8 23 23 23 26 26 26 29 29 29 32 32 32 35 35 35 23 23 23 26 26 26 29 29 29 32 32 32 35 35 35 - - - - - - - - - - - - 207 y = 3.98E1x - 3.43E2 r2 = 0.097 *** y = 4.99E2x - 9.87E2 * r2 = 0.107 y = 4.87E2x - 6.94E3 r2 = 0.037 *** e l o e n C 8 i , 1 n o i t a r t n e c n o c ) M D 1 - g m · g n ( l o o l a n L i n o i t a r t n e c n o c ) M D 1 - g m · g n ( l o n e g u E n o i t a r t n e c n o c ) M D 1 - g m · g n ( l o c i v a h c l y h t e n o i t a r t n e c n o c M ) M D 1 - g m · g n ( 1400 1200 1000 800 600 400 200 0 8000 6000 4000 2000 0 12000 10000 8000 6000 4000 2000 0 14000 12000 10000 8000 6000 4000 2000 0 A B C D 34 36 38 Figure V-1. Concentrations [ng·mg‒1 dry mass (DM)] of 1,8 cineole (A), linalool (B), eugenol 22 24 26 30 28 32 MDT (ºC) 208 Figure V-1 (cont’d). (C), and methyl chavicol (D) of sweet basil (Ocimum basilicum ‘Nufar’) grown at mean daily temperatures (MDT) from 23 to 36 °C for three weeks. Symbols represent the mean of 15 plants ± SE. Lines represent linear regression. * and *** indicate significant at P ≤ 0.05 or 0.001, respectively. 209 g n i t a r t r e k L n a e M i g n i t a r t r e k i L n a e M g n i t a r t r e k i L n a e M g n i t a r t r e k L n a e M i 7.6 7.4 7.2 7.0 6.8 6.6 6.4 1.0 7.2 7.0 6.8 6.6 6.4 6.2 6.0 5.8 1.0 7.8 7.6 7.4 7.2 7.0 6.8 6.6 6.4 1.0 4.0 3.8 3.6 3.4 3.2 3.0 1.0 Appearance ab a A ab Texture a ab B ab Color a a C a ab ab ab Bitterness/Sweetness ab D ab ab b b b b a 24 26 28 30 32 34 36 MDT (ºC) 210 Figure V-2 (cont’d). Appearance (A), texture (B), color (C), and bitterness/sweetness (D) of Figure V-2 (cont’d). sweet basil (Ocimum basilicum ‘Nufar’) grown at mean daily temperatures (MDT) from 25 to 35 °C for three weeks, based on a 9-point Likert scale ranging from dislike extremely (1) to like extremely (9). Means not followed by the same letter are significantly different by Tukey’s honestly significant difference test (P < 0.05). Each bar represents 86 responses ± SD. 211 Figure V-3. Principal component analysis (PCA) showing biplot differentiation of sweet basil (Ocimum basilicum ‘Nufar’) grown at mean daily temperatures (MDT) from 23 to 36 °C for three weeks, based on consumer sensory preferences (n = 86) and the concentration of two terpenoids and two phenylpropanoids (n = 15). 212 s s e n t e e w s / s s e n r e t t i B * e t s a t r e t f A e z i s f a e L e r u t x e T * r o l o C * l o c i v a h c l y h t e M l o o l a n i L l o n e g u E l l a r e v O e l o e n i C 8 1 , Appearance Aroma Flavor Texture Color Leaf size Bitterness/sweetness Aftertaste 1,8 Cineole Linalool Methyl chavicol Eugenol Overall a m o r A r o v a l F - - e c n a r a e p p A - * * * - * ** * Correlation -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 * - * * * * - ** - * ** * - - - * * - - - ** * * * Figure V-4. Heat map illustrating the correlation between volatile organic compounds (VOCs) and sensory preference characteristics of sweet basil (Ocimum basilicum ‘Nufar’) grown at mean daily temperatures (MDT) from 23 to 36 °C for three weeks. Blue and red represent negative and positive correlations, respectively. Asterisks indicate significant correlations based on Pearson’s correlation *, P < 0.05; **, P < 0.01. 213 LITERATURE CITED 214 LITERATURE CITED Adams, S.R., S. Pearson, and P. Hadley. 1997. The effects of temperature, photoperiod and light integral on the time to flowering of pansy cv. Universal Violet (Viola × wittrockiana Gams.). Annals of Botany, 80(1):107‒112. Arena, E., N. Guarrera, S. Campisi, and C.N. Asmundo. 2006. Comparison of odour active compounds detected by gas-chromatography–olfactometry between hand-squeezed juices from different orange varieties. Food Chemistry, 98(1):59–63. Bernacchi, C.J., A.R. Portis, H. Nakano, S. Von Caemmerer, and S.P. Long. 2002. Temperature response of mesophyll conductance. Implications for the determination of Rubisco enzyme kinetics and for limitations to photosynthesis in vivo. Plant Physiology, 130(4):1992‒1998. Boldt, J.K., M.H. Meyer, and J.E. Erwin. 2014. Foliar anthocyanins: A horticultural review. Horticultural Reviews. 42;209‒252. Busch, G., E. Bayer, A. Gunarathne, S. Hölker, S. Iweala, K. Jürkenbeck, D. Lemken, C. Mehlhose, M. Ohlau, A. Risius, and C. Rubach. 2020. Purchasing and nutritional behavior and resilience of the nutritional system from a population perspective: Results of a study during the Corona pandemic in April 2020 (No. 2003). Georg-August University of Göttingen, Department of Agricultural Economics and Rural Development. Carvalho, S.D., M.L. Schwieterman, C.E. Abrahan, T.A. Colquhoun, and K.M. Folta. 2016. Light quality dependent changes in morphology, antioxidant capacity, and volatile production in sweet basil (Ocimum basilicum). Frontiers in Plant Science, 7:1328. Chang, X., P. Alderson, and C. Wright, 2005. Effect of temperature integration on the growth and volatile oil content of basil (Ocimum basilicum L.). The Journal of Horticultural Science and Biotechnology, 80(5):593‒598. Chang, X., P.G. Alderson, T.A. Hollowood, L. Hewson, and C.J. Wright. 2007. Flavour and aroma of fresh basil are affected by temperature. Journal of the Science of Food and Agriculture, 87(7):1381–1385. Choi, N.E. and J.H. Han. 2015. How Flavor Works: The Science of Taste and Aroma. John Wiley & Sons, Chichester, UK. Cseke, L.J., A. Kirakosyan, P.B. Kaufman, S. Warber, J.A. Duke, and H.L. Brielmann. 2016. Natural products from plants. CRC press, Boca Raton, FL. De Vincenzi, M., M. Silano, A. De Vincenzi, F. Maialetti, and B. Scazzocchio. 2002. Constituents of aromatic plants: Eucalyptol. Fitoterapia, 73(3):269–275. 215 Gomez, C., C.J. Currey, R.W. Dickson, H. Kim, R. Hernandez, N.C. Sabeh, R.E. Raudales, R.G. Brumfield, A. Laury-Shaw, A.K. Wilke, R.G. Lopez, and S.E. Burnett. 2019. Controlled environment food production for urban agriculture. HortScience, 54(9):1448–1458. Hälvä, S., L.E. Craker, J.E. Simon, and D.J. Charles. 1993. Growth and essential oil in dill, Anethum graveolens L., in response to temperature and photoperiod. Journal of Herbs, Spices, and Medicinal Plants, 1(3):47–56. Iijima, Y., R. Davidovich-Rikanati, E. Fridman, D.R. Gang, E. Bar, E. Lewinsohn, and E. Pichersky. 2004. The biochemical and molecular basis for the divergent patterns in the biosynthesis of terpenes and phenylpropenes in the peltate glands of three cultivars of basil. Plant Physiology, 136(3):3724–3736. Kanellos, E.A.G. and S. Pearson. 2000. Environmental regulation of flowering and growth of Cosmos atrosanguineus (Hook.) Voss. Scientia horticulturae, 83(3–4):265–274. Karlsson M.G., R.D. Heins, J.E. Erwin.1988. Quantifying temperature controlled leaf unfolding rates in Lilium longiflorum Thunb. Journal of the American Society for Horticultural Science, 113:70–74. Keeling, P.L., R. Banisadr, L. Barone, B.P. Wasserman, and G.W. Singletary. 1994. Effect of temperature on enzymes in the pathway of starch biosynthesis in developing wheat and maize grain. Functional Plant Biology, 21(6):807–827. Kolodinsky, J., M. Sitaker, L. Chase, D. Smith, and W. Wang. 2020. Food systems disruptions. Journal of Agriculture, Food Systems, and Community Development, 9:1–4. Leyva, A., J.A. Jarillo, J. Salinas, and J.M. Martinez-Zapater. 1995. Low temperature induces the accumulation of phenylalanine ammonia-lyase and chalcone synthase mRNAs of Arabidopsis thaliana in a light-dependent manner. Plant Physiology, 108(1):39–46. Naim, M., S. Nir, A.I. Spielman, A.C. Noble, I. Peri, S. Rodin, and M. Samuelov-Zubare. 2002. Hypothesis of receptor-dependent and receptor-independent mechanisms for bitter and sweet taste transduction: implications for slow taste onset and lingering aftertaste. p. 2- 17. In: P. Given and D. Paredes (eds.) Chemistry of taste: Mechanisms, Behaviors, and Mimics. American Chemical Society, Washington, D.C. Pramuk, L.A. and E.S. Runkle, 2005. Modeling growth and development of celosia and impatiens in response to temperature and photosynthetic daily light integral. Journal of the American Society for Horticultural Science, 130(6):813–818. Pogany, D., C.L. Bell, and E.R. Kirch. 1968. Composition of oil of sweet basil (Ocimum basilicum L.) obtained from plants grown at different temperatures. Perfumery Essential Oil Record, 59:858–865. Sage, R.F. and D.S. Kubien. 2007. The temperature response of C3 and C4 photosynthesis. Plant, Cell, and Environment, 30(9):1086–1106. 216 Santos, A., G. Chierice, K. Alexander, A. Riga, and E. Matthews. 2009. Characterization of the raw essential oil eugenol extracted from Syzygium aromaticum L. Journal of Thermal Analysis and Calorimetry, 96(3):821–825. Schilmiller, A., F. Shi, J. Kim, A.L. Charbonneau, D. Holmes, A.D. Jones, and R.L. Last. 2010. Mass spectrometry screening reveals widespread diversity in trichome specialized metabolites of tomato chromosomal substitution lines. The Plant Journal, 62(3):391–403. Sheen, L.Y., Y.H.T. Ou, and S.J. Tsai. 1991. Flavor characteristic compounds found in the essential oil of Ocimum basilicum L. with sensory evaluation and statistical analysis. Journal of Agricultural and Food Chemistry, 39(5):939–943. Simon, J.E., M.R. Morales, W.B. Phippen, R.F. Vieira, and Z. Hao. 1999. Basil: A source of aroma compounds and a popular culinary and ornamental herb. Perspectives on New Crops and New Uses, 16:499–505. Tollenaar, M., T.B. Daynard, R.B. Hunter. 1979. Effect of temperature on rate of leaf appearance and flowering date in maize. Crop Science, 19(3):363–366. Walters, K.J., B.K. Behe, C.J. Currey, and R.G. Lopez. 2020. Historical, current, and future perspectives for controlled environment hydroponic food crop production in the United States. HortScience, 55(6):758–767. Warner, R.M., 2020. Differential temperature sensitivity of flowering time and crop quality parameters of 20 seed-propagated petunia cultivars. HortScience, 55(3):362–367. Weisshaar, B. and G.I. Jenkins. 1998. Phenylpropanoid biosynthesis and its regulation. Current Opinion in Plant Biology, 1(3):251–257. 217 CHAPTER 6 SOLE-SOURCE RADIATION INTENSITY AND CARBON DIOXIDE CONCENTRATION DURING BASIL SEEDLING PRODUCTION INFLUENCE SUBSEQUENT YIELD AND FLAVOR COMPOUND CONCENTRATION DURING GREENHOUSE PRODUCTION 218 Sole-source radiation intensity and carbon dioxide concentration during basil seedling production influence subsequent yield and flavor compound concentration during greenhouse production Kellie J. Walters and Roberto G. Lopez* This work was supported by Michigan State University AgBioResearch (including Project GREEEN GR19-019), the USDA National Institute of Food and Agriculture Hatch project MICL02472 and The Fred C. Gloeckner Foundation. We gratefully acknowledge Nate DuRussel for assistance, JR Peters for fertilizer, Grodan for substrate, Fluence by OSRAM for LED fixtures, and Hydrofarm for hydroponic production systems. The use of trade names in this publication does not imply endorsement by Michigan State University of products named nor criticism of similar ones not mentioned 219 Additional Index Words: 1,8 cineole, daily light integral, estragole, eucalyptol, eugenol, hydroponic, linalool, methyl chavicol, Ocimum basilicum, propagation Abstract. Under indoor sole-source lighting, radiation intensity and carbon dioxide (CO2) concentration can be precisely controlled to manipulate plant growth and development. Due to increased plant densities during seedling production, fewer inputs per plant are required, creating the potential to increase production efficiency. Therefore, the objectives of this research were to: 1) quantify the extent radiation intensity and CO2 concentration under sole-source lighting influence morphology and yield of sweet basil (Ocimum basilicum) seedlings, and 2) determine if differences in morphology, yield, and volatile organic compound (VOC) concentration persist after transplant in a common environment. Sweet basil ‘Nufar’ seeds were sown in rockwool cubes and placed in growth chambers with target CO2 concentrations of 500 or 1,000 µmol·mol‒ 1. Light-emitting diodes (LEDs) provided 20:40:40 blue:green:red radiation ratios (%), a red:far- red ratio of 13:1, and target radiation intensities of 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD) for 16 h to create daily light integrals (DLIs) of 6, 12, 23, or 35 mol·m‒2·d‒1. After two weeks, seedlings were transplanted into deep flow technique hydroponic systems in a greenhouse with an average daily temperature of ~23 °C and DLI of ~14 mol·m‒2·d‒1. At transplant and three weeks after transplant (harvest), height, leaf area, stem diameter, and shoot fresh and dry mass were recorded. Node and branch number along with concentrations of terpenoids 1,8 cineole and linalool, and phenylpropanoids eugenol and methyl chavicol were measured at harvest. Carbon dioxide and radiation intensity interacted to influence seedling height, leaf area, and stem width and harvest height and stem width, though the effects 220 of CO2 concentration were less pronounced than that of radiation intensity. As radiation intensity during seedling production increased from 100 to 600 µmol·m‒2·s‒1, basil seedlings were 38% (0.6 cm) taller, had a 713% (9.6 cm3) larger leaf area, and had 65% (0.8 mm) thicker stems; at harvest, plants were 5 cm taller (24% increase), had 2 more branches (56% increase), 1 more node (28% increase), 1.3-mm thicker stems (22% increase), and weighed 25 g more when fresh (80% increase) and 2 g more when dry (80% increase). Additionally, after growing in a common environment for three weeks, eugenol concentration was greater in plants grown under a PPFD of 600 µmol·m‒2·s‒1 as seedlings compared to lower intensities. Therefore, increasing radiation intensity during sole-source lighting seedling production can carry over to increase subsequent yield and eugenol concentration during finished production. Introduction Currently, the United States (U.S.) demand for culinary herbs exceeds domestic production, even with controlled environment (CE) production area increasing by 134% and the number of operations increasing by 62% from 2009 to 2014 (USAID, 2014; USDA, 2015). Although greenhouses are an example of CEs, it is often difficult to maintain consistent temperatures, radiation levels, and carbon dioxide (CO2) concentrations throughout the year (Korczynski et al., 2002). This in turn makes consistent year-round production of food crops challenging. Indoor plant factories and vertical farms can be more precisely controlled, especially for difficult to grow (i.e., tissue-culture transplants) and high-value young plants, to improve uniformity and quality while reducing production time and losses. However, the energy cost of greenhouse heating, lighting, and fans, etc. is often less than that incurred with sole- source lighting, heating, ventilation, and air conditioning used in indoor production systems 221 (Gomez et al., 2019). Recent advances and increased efficacy of light-emitting diode (LED) fixtures have made sole-source lighting and indoor production more feasible for certain types of production (Kozai, 2013). With more expensive capital and operating costs for indoor production, production of short duration and high-density crops is potentially more profitable (Gibson et al., 2020). For example, a 200-cell tray of culinary herb seedlings can be produced in two weeks in the same CE area that three to 20 plants can be grown and harvested in three weeks. If the higher operating and capital cost can be spread across a larger number of plants with a shorter production duration, the cost per plant is less. Under indoor sole-source lighting, radiation intensity, temperature, and CO2 concentration can be precisely controlled to improve growth, development, and volatile organic compound (VOC) concentrations. However, there is currently limited information on physiological and biochemical responses of culinary herbs to varying radiation intensities and CO2 concentrations under sole-source lighting. With the high input costs during indoor production, maximizing photosynthesis and biomass production are often correlated and integral for production optimization (Gomez et al., 2019). In C3 plants such as basil, biomass production is largely determined by radiation intensity and CO2 concentration, as these two parameters influence carbon accumulation through photosynthesis and photorespiration (Sharkey, 1985). There are three main limitations to photosynthesis: the supply or utilization of CO2, radiation, and phosphate (Sharkey, 1985). The limiting factor in photosynthesis depends partly on the environment; for example, CO2-limited photosynthesis does not respond to increased radiation (Blackman 1905; Sharkey, 1985). If photosynthesis is radiation-limited, increasing CO2 concentration will not result in large increases in photosynthesis. However, radiation-limited photosynthesis is still sensitive to CO2 222 and when CO2 concentrations are increased, less oxygenation and photorespiration occurs (Ogren and Bowes, 1971). Research towards optimizing CO2 and radiation intensity or daily light integral (DLI) in CEs has been occurring for decades. Wittwer and Robb (1964) grew lettuce ‘Grand Rapids’ (Lactuca sativa), tomato ‘Michigan-Ohio Hybrid’ (Solanum lycopersicum), and cucumber ‘Burpee Hybrid’ (Cucumis sativus) under 73 or 218 µmol·m–2·s–1 photosynthetic photon flux density (PPFD; 3.2 or 9.4 mol·m–2·d–1 DLI), and 400 or 1,000 µmol·mol–1 CO2; as both radiation and CO2 concentration increased, fresh and dry mass increased. Elevated CO2 concentration is most efficient at high radiation intensities (Chermnykh and Kosobrukhov, 1987). For example, in cucumber ‘Moskovsky Teplichnyi’, increasing CO2 concentration from ambient to 1,000 µmol·mol–1 had an insignificant effect on photosynthesis when radiation intensity was ~60 to ~320 µmol·m–2·s–1 but increased photosynthetic rate when radiation was increased to ~480 µmol·m–2·s–1(Chermnykh and Kosobrukhov, 1987). Thus, the benefits of elevated CO2 concentrations are generally realized at high radiation intensities. Researchers have proposed elevating CO2 concentration during seedling propagation is the more advantageous compared to other stages of production due to their rapid vegetative growth (Craver, 2018; Thomas et al., 1975; Tremblay and Gosselin, 1998). Tobacco (Nicotiana tabacum) seedling growth rate was greater when grown in 1,000 compared to 400 µmol·mol–1 CO2, and overall, seedling growth rate was 3-fold greater than post-transplant (Thomas et al., 1975). Increasing CO2 concentration from 320 to 750 µmol·mol–1 during the propagation of 96 tomato cultivars increased dry weight by a factor of 2.3 (Lindhout and Pet, 1990). Similarly, cucumber ‘Burpee Hybrid’, lettuce ‘Grand Rapids’, and tomato ‘Michigan-Ohio’ seedling dry 223 mass increased 2 to 4.6 times as CO2 concentration was increased from 400 to 2,000 µmol·mol–1 (Krizek et al., 1974). In addition to biomass production, crop quality, especially improved flavor, is an integral goal of crop production. CE growers have indicated a need for research on adjusting the growing environment to improve crop flavor (Goodman and Minner, 2018; Walters et al., 2020). Many VOCs contribute to basil flavor, including phenylpropanoids and terpenoids. Eugenol is a phenylpropanoid that contributes a clove-like flavor and aroma, while methyl chavicol (estragole) is more anise-like (Santos et al., 2009; Simon et al., 1999). The terpenoid linalool can be described as floral or spicy (Arena, et al., 2006) or reminiscent of the cereal “Fruit Loops®”. 1,8 Cineole, another terpenoid, contributes an aroma and flavor analogous to eucalyptus (Eucalyptus globulus; De Vincenzi et al., 2002). Daily light integral and CO2 concentration influence secondary metabolite concentrations, including VOCs. Dou et al. (2018) determined that increasing DLI from 9.3 to 17.8 mol·m–2·d–1 not only increased basil fresh mass, net photosynthesis, and leaf area and thickness, but also increased anthocyanin, phenolic, and flavonoid concentrations. Chang et al. (2008) reported that as the DLI delivered for two weeks to young seedlings increased from 5.3 to 24.9 mol·m–2·d–1, total VOC content of basil increased. In particular, the relative content of eugenol and linalool increased ~300% and ~400%, respectively, while methyl eugenol relative content decreased by ~80%. Carbon dioxide can also influence secondary metabolite production. For example, linalool, a compound of interest in basil, is also present in strawberry (Fragaria ananassa; Wang and Bunce, 2004). An increase in CO2 concentration from ~350 to ~950 µmol·mol–1 increased linalool concentration during strawberry production (Wang and Bunce, 2004). By investigating the individual and combined influences of DLI and CO2 on VOC content 224 and concentration, CE growers can work toward optimizing growing conditions that increase plant quality (VOC content and concentration) and yield (fresh mass). Although technological advances have made indoor plant production more economically feasible, a better understanding of how to leverage environmental controls to improve crop productivity, quality, and energy efficiency is needed (Banerjee and Adenaeuer, 2014). While researchers have been mainly focused on investigating the influence of environmental variables on finished-stage crops, the potential to improve high-density young plant production creates an opportunity to spread potentially greater input costs over a larger number of plants. Therefore, the objectives of this research were to: 1) quantify the extent radiation intensity and CO2 concentration during seedling production influence yield, 2) determine if physiological and morphological differences remain present after transplant in a greenhouse, and 3) determine if differences in VOC concentration due to radiation intensity at the seedling stage remain present through harvest in a common greenhouse environment. Our hypotheses were that 1) growth would increase as radiation intensity and CO2 increased, 2) there would be a positive interactive effect between CO2 concentration and radiation intensity, where the effects of elevated CO2 concentration would be more pronounced when the radiation intensity was higher, and 3) increased VOC concentrations at transplant would not persist through finishing in a common environment due to dilution during rapid growth. Materials and Methods Seedling production. Sweet basil ‘Nufar’ (Johnny’s Selected Seeds, Fairfield, ME) was selected based on disease resistance and comparatively high yield results from Walters and Currey (2015). Seeds 225 were sown two per cell in stone wool cubes (2.5 × 2.5 × 4 cm, AO plug; Grodan, Roermond, Netherlands) and 200-cell flats were placed in one of two walk-in growth chambers (Hotpack environmental room UWP 2614-3; SP Scientific, Warminster, PA) on 7 Aug. 2017, 10 Nov. 2017, and 22 Jan. 2018. Seeds and seedlings were grown and the environmental conditions were controlled and monitored as reported in Chapter 4. Light-emitting diodes (LEDs) provided 20:40:40 blue:green:red radiation ratios (%), a red:far-red ratio of 13:1, and target radiation intensities of 100, 200, 400, or 600 µmol·m‒2·s‒1 photosynthetic photon flux density (PPFD) for a 16-h photoperiod to create daily light integrals (DLIs) of 6, 12, 23, or 35 mol·m‒2·d‒1. Fixture density and hanging height were adjusted to achieve target radiation intensities. Radiation intensity and spectrum were measured at four corners and in the center of the seedling flat with a spectroradiometer (PS-200; StellarNet, Inc., Tampa, FL) to quantify the intensities and spectrum across the growing area (Fig. VI-1, Table VI-1). Target carbon dioxide concentrations of 500 and 1000 µmol·mol–1 were maintained by injecting compressed CO2 to increase concentrations and scrub CO2 using soda lime scrubber (Environmental Growth Chambers) to decrease concentrations. Concentrations were measured with a CO2 sensor (GM86P; Vaisala, Helsinki, Finland) and logged by a C6 Controller (Environmental Growth Chambers) every 5 s (Table VI-1). Finished plant production. Two weeks after sowing, 17 seedlings were transplanted into 0.9-m-wide by 1.8-m-long deep-flow hydroponic systems (Active aqua premium high-rise flood table; Hydrofarm, Petaluma, CA) in a glass-glazed greenhouse. Baskets holding the seedlings were placed in 4-cm- diameter holes, 20-cm apart, in 4-cm thick extruded polystyrene foam floating on the nutrient solution. The nutrient solution consisted of reverse osmosis water supplemented with 12N-1.8P- 226 13.4K water-soluble fertilizer (RO Hydro FeED; JR Peters, Inc.) and MgSO4 providing twice the concentrations reported previously. Electrical conductivity (EC) and pH were measured (HI991301 Portable Waterproof pH/EC/TDS Meter; Hanna Instruments, Woonsocket, RI) and adjusted to 1.56 mS·cm–1 and 6.0, respectively, by adding fertilizer, reverse osmosis water, potassium bicarbonate, or sulfuric acid. Air pumps (Active aqua 70 L·min–1 commercial air pump; Hydrofarm) and air stones (Active aqua air stone round 4”x1”; Hydrofarm) were used to increase dissolved oxygen concentrations. The 16-h (0600 to 2200 HR) photoperiod consisted of natural photoperiods (lat. 43º N) and day-extension lighting from high-pressure sodium (HPS) lamps providing a supplemental PPFD of ~75 µmol·m–2·s–1 to achieve target DLIs of 13 to 17 mol·m–2·d–1. Target average daily temperature was a constant 23 °C. Exhaust fans, evaporative-pad cooling, radiant steam heating, and supplemental lighting was controlled by an environmental control system (Integro 725; Priva North America, Vineland Station, ON, Canada). Shielded and aspirated 0.13-mm type E thermocouples (Omega Engineering) measured air temperature, infrared thermocouples (OS36- 01-T-80F; Omega Engineering) measured leaf temperature, and quantum sensors (LI-190R Quantum Sensor; LI-COR Biosciences) placed at canopy height recorded PPFD. Every 15 s, a CR-1000 datalogger (Campbell Scientific) collected environmental data and hourly means were recorded (Table VI-2). Growth, development, and VOC data collection and analysis. At transplant and three weeks after transplant (harvest), height from the substrate surface to the meristem, leaf area of two (seedling) or four (harvest) most recently fully expanded leaves (measured with LI-300; LI-COR Biosciences), stem diameter (harvest reps 2 and 3) at the base, and shoot fresh mass were recorded at transplant and three weeks after transplant (harvest). 227 Additionally, the number of branches >2.5 cm and node number (rep 2 and 3) were recorded at harvest. Tissue was placed in a forced-air oven maintained at 75 ºC for at least 3 d, weighed, and dry mass was recorded. Three weeks after transplant, the two most recent, fully mature leaves of five plants from each treatment were detached, frozen, and stored at -20 °C until gas chromatography mass spectrometry (GCMS) analysis as reported in Chapter 4 from a method derived from Schilmiller et al. (2010). Statistical design and analysis. The seedling portion of this experiment was organized the same as Chapter 4, a split-plot design with two CO2 concentrations (two growth chambers) as the main factor and four radiation intensities as the sub factor. Finished greenhouse production was organized in a randomized complete block design with seedlings from the growth chamber blocked by treatment. The experiment was completed thrice over time for growth and development analysis, and twice in time for GCMS analysis (reps 2 and 3). Analysis of variance and Tukey’s honestly significant difference tests were performed using JMP (version 12.0.1, SAS Institute Inc., Cary, NC). When interactions were not significant, data were pooled. Linear and quadratic regression analyses were conducted using Sigma Plot (version 11.0, Systat Software Inc., San Jose, CA). Results Seedlings. Radiation intensity but not CO2 concentration influenced fresh and dry mass (Figs. VI-2B and D). Fresh mass increased linearly from 0.134 to 0.515 g (284%) and dry mass increased quadratically from 0.009 to 0.062 g (589%) as radiation intensity increased from 100 to 600 µmol·m–2·s–1. 228 Chamber CO2 concentration and radiation intensity interacted to affect height, leaf area, and stem width. As radiation intensity increased, height, leaf area, and stem width generally increased (Figs. VI-2A, C, and E). As radiation intensity increased from 100 to 600 µmol·m–2·s– 1, basil seedlings were 38% (0.6 cm) taller, had a 713% (9.6 cm3) larger leaf area, and 65% (0.8 mm) thicker stems. Seedlings grown under 100 µmol·m–2·s–1 were a similar height (1.6 cm) when grown at 500 or 1,000 µmol·mol–1 CO2. Plants grown under 200 µmol·m–2·s–1 were 0.3 cm taller when grown at 1,000 than 500 µmol·mol–1 CO2, and plants grown under 600 µmol·m–2·s–1 were 0.2 cm shorter when grown at 1,000 than 500 µmol·mol–1 CO2 (Fig. VI-2A). Similarly, when grown under lower radiation intensities (100 or 200 µmol·m–2·s–1), stem width was 0.1 mm greater with a CO2 concentration of 1,000 than 500 µmol·mol–1. However, when plants were grown under higher intensities (600 µmol·m–2·s–1), stem width was 0.1 mm greater at 500 µmol·mol–1 compared to 1,000 µmol·mol–1 CO2 (Fig VI-2E). Under a radiation intensity of 200 µmol·m–2·s–1, leaf area of seedlings was 3.1 cm3 larger at 1,000 µmol·mol–1 than at 500 µmol·mol–1 CO2, while the leaf area was 2.3 cm3 smaller under 600 µmol·m–2·s–1 (Fig VI-2C). Harvest. In general, radiation intensity during the seedling stage influenced height, branch and node number, stem width, and fresh and dry mass at harvest, three weeks after transplant (Figs. VI-3A–F). Plants grown under a radiation intensity of 600 µmol·m–2·s–1 as seedlings, were 5 cm taller (24% increase), and had 2 more branches (56% increase), 1 more node (28% increase), 1.3- mm thicker stems (22% increase), and 25 g more fresh mass (80% increase), and 2 g more dry mass (80% increase) at harvest compared to plants grown under 100 µmol·m–2·s–1 as seedlings. Leaf area of the four leaves measured at harvest was not affected by radiation intensity or CO2 concentration during the seedling stage (data not shown). 229 Similar to the seedling stage, radiation intensity and CO2 concentration interacted to affect height and stem width at harvest (Figs. VI-3A and C). Plants grown under 100 or 200 µmol·m–2·s–1 during the seedling stage had similar height and stem width at harvest regardless of CO2 concentration. Basil grown under a PPFD of 600 µmol·m–2·s–1 as seedlings were 2.4 cm (10%) taller and had 1.4 mm (22%) thicker stems when grown at 500 µmol·mol–1 compared to 1,000 µmol·mol–1 CO2 (Figs. VI-3A and C). Finished volatile organic compound concentrations. After seedlings were transplanted into a common greenhouse environment and grown for three weeks, there was no difference in linalool, 1,8 cineole, or methyl chavicol concentrations due to radiation intensity or CO2 concentration provided during the seedling stage (Figs. VI-4A, B, and D). However, as the radiation intensity during seedling production increased, an overall quadratic increase in eugenol concentration persisted through finishing (Fig. VI-4C). There were minimal differences in eugenol concentration among plants grown under a PPFD of 100, 200, or 400 µmol·m‒2·s‒1 during the seedling stage, three weeks after transplant. Increasing the radiation intensity to 600 µmol·m‒2·s‒1 during the seedling stage increased eugenol concentration 44% to 183% (1,327 to 1,946 ng·mg‒1 dry mass) compared to the lower radiation intensity treatments. There was no effect of CO2 concentration on VOCs. Discussion Increased radiation intensity increased growth and morphological attributes. Increasing radiation intensity or DLI up to a saturating value increases biomass production. Walters and Currey (2018) reported that sweet basil ‘Nufar’ fresh mass increased 144% as DLI increased from 7 to 15 mol·m‒2·d‒1. Increasing DLI from 9.3 to 17.8 mol·m‒2·d‒1 230 during sweet basil ‘Improved Genovese Compact’ production increased fresh mass 78% (Dou et al., 2018), and increasing DLI from 2 to 20 mol·m‒2·d‒1 increased sweet basil ‘Nufar’ fresh mass 24-fold (Litvin, 2019). However, none of these studies increased DLI above 20 mol·m‒2·d‒1. The goal of Beaman et al. (2008) was to determine the radiation intensity that led to the highest sweet basil biomass production. Sweet basil ‘Nufar’ shoot fresh mass increased 39% as radiation intensity increased from a PPFD of 300 to 500 µmol·m‒2·s‒1 (DLIs of 17.3 to 28.8 mol·m‒2·d‒1), while fresh mass was similar among plants grown under 500 and 600 µmol·m‒2·s‒1 (28.8 and 34.6 mol·m‒2·d‒1). Though these previously mentioned studies were conducted at harvest on plants in the finished stage of production, our study with seedlings concurs; increasing radiation intensity from 100 to 600 µmol·m‒2·s‒1 (DLI of 5.8 to 34.6 mol·m‒2·d‒1) increased seedling fresh mass by 284% (Fig. VI-2D). Our results confirm that regardless of production stage, radiation intensity during sweet basil production can be increased up to 600 µmol·m‒2·s‒1 (34.6 mol·m‒ 2·d‒1) to increase fresh mass. CO2 concentration did not influence mass. Contrary to our hypothesis, CO2 concentration did not influence fresh or dry mass at transplant or harvest (Figs. VI-2B and D, VI-3E and F). While most research illustrates increased CO2 concentrations can increase biomass, there are a few reasons plants may not respond to elevated CO2 concentrations. Plants can become acclimated to elevated CO2 concentrations, with prolonged exposure becoming inhibitory to photosynthesis; however, the extent and presence of acclimation or negative effects are species- and potentially, production stage-dependent (Craver, 2018; Sage et al., 1989). For example, Sage et al. (1989) reported that long-term elevated (900 to 1000 µmol·mol–1) CO2 negatively affected the photosynthetic rate C3 plants such as kidney bean ‘Linden’ (Phaseolus vulgaris), eggplant (Solanum melongena), and cabbage (Brassica oleracea), 231 but the photosynthetic rate of C3 plant lambsquarters (Chenopodium album) increased. While the main benefit of elevated CO2 is the favoring of carboxylation activity over oxygenation activity of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco), initial increased photosynthetic rates can cause excess carbohydrate production and result in feedback inhibition, thus a reduction in photosynthesis (Arp, 1997; Craver, 2018; Sage et al., 1989). Additionally, in some species, elevated CO2 decreased Rubisco content and/or activity (Craver, 2018; Sage et al., 1989), induced stomatal closure, and/or decreased stomatal density (Ainsworth and Rogers, 2007; Craver, 2018; Sage et al., 1989). The lack of CO2 effect on biomass could also be due to the concentration being near or above the CO2 saturation point. Previous research determined that increasing CO2 concentrations from 360 to 620 µmol·mol–1 increased ~4-week old basil (cultivar not reported) fresh mass 40% when grown under a radiation intensity of 150 µmol·m‒2·s‒1 (8.6 mol·m‒2·d‒1; Al Jaouni et al., 2018). We hypothesize that the increased mass with increased CO2 concentration reported by Al Janouni et al. (2018) was due to CO2 concentrations being below the CO2 saturation point of 729 µmol·m‒2·s‒1 determined by Park et al. (2016) when sweet basil (cultivar not reported) was acclimated to 20 °C, 400 µmol·mol–1 CO2, and a radiation intensity of 150 µmol·m‒2·s‒1 (9.7 mol·m‒2·d‒1). Additionally, as CO2 concentration approaches the saturation point, increases in mass are attenuated. Since we utilized 500 and 1,000 µmol·mol–1 CO2, 500 µmol·mol–1 may have been too similar to the CO2 saturation point and 1,000 µmol·mol–1 may have been above the saturation point, resulting in no discernable difference in mass. If we had maintained CO2 concentrations below 500 µmol·mol–1, differences may have occurred; however, this hypothesis would have to be tested. 232 Another contributing factor could be that CO2 utilization may be limited by environmental factors including temperature. As temperature increases, Rubisco has a higher affinity for oxygen; therefore, the positive influence of elevated CO2 on photosynthesis increases as temperature increases from approximately 20 to 35 °C, though the effect is species-dependent (Berry and Raison, 1981; Chermnykh and Kosobrukhov, 1987; Long, 1991). In this study, basil was grown at ~23 °C. If temperatures had been higher and closer to the optimal temperature for sweet basil growth and development of 32 to 35 °C (Chapter 3), the elevated CO2 concentration may have been more likely to have had an effect on fresh mass. CO2 concentration influenced morphology. The increased height, leaf area, and stem width due to lower radiation intensities (100 to 400 µmol·m‒2·s‒1; Figs. VI-2A, C, and E) with elevated CO2 concentration were likely due to differing biomass partitioning, since neither fresh nor dry mass were affected (Arp, 1991). In wheat ‘WW15’ (Triticum aestivum), leaf area index increased at elevated CO2 concentrations under lower radiation conditions (Gifford, 1977). However, there are conflicting reports on the effect of elevated CO2 on leaf area. In separate experiments under different environmental conditions, leaf area of tomato ‘Minibelle’ increased as CO2 concentration increased (Hurd and Thornley, 1974), however, tomato ‘Findon Cross’ leaf area was unaffected (Besford et al., 1990). The lower height, leaf area, and stem width when basil seedlings were under a radiation intensity of 600 µmol·m‒2·s‒1 and elevated CO2 was counterintuitive (Figs. VI-2A, C, and E). It is well documented that as radiation intensity increases, leaf thickness increases (Dou et al., 2018). It could be possible that increased tissue thickness can impact plant responses to elevated CO2, however, additional morphological and physiological data is needed to confirm or reject the hypothesis. 233 Seedling production conditions influence basil yield and quality at harvest. Increasing radiation intensity from 100 to 600 µmol·m‒2·s‒1 (5.8 to 34.6 mol·m‒2·d‒1) increased seedling fresh mass 284% (Fig. VI-2D), and an 80% increase in fresh mass yield persisted through finishing in a common environment (Fig. VI-3C). Radiation intensity during propagation of floriculture crops can have a profound effect on subsequent growth and development. For example, increasing DLI from 4.1 to 14.2 mol·m‒2·d‒1 during seedling production hastened flowering and reduced shoot dry weight at flower for celosia (Celosia argentea var. plumosa), impatiens (Impatiens walleriana), French marigold (Tagetes), and pansy (Viola; Pramuk and Runkle, 2005). However, the reduction in dry mass can be primarily attributed to earlier flowering and thus, a shorter production duration. In the present study, plants were not grown until anthesis, but were harvested at the same time. Therefore, the influence of radiation intensity observed at transplant persisted, but was attenuated. Similar to the floriculture studies, development, including node and branch number, was hastened by increased radiation intensity during propagation in our study. In addition to increased yield at harvest, and contrary to our hypothesis, the higher eugenol concentrations in seedlings grown under high radiation intensities (Chapter 4) persisted at harvest (Fig. VI-4C). In Chapter 4, we observed that increasing radiation intensity from 100 to 600 µmol·m‒2·s‒1 (5.8 to 34.6 mol·m‒2·d‒1) during basil seedling production increased 1,8 cineole, linalool, and eugenol concentrations. In a study investigating the influence of radiation quality on basil VOCs, researchers suggested that specific light treatments during germination and early seedling development “may install a particular developmental/metabolic pattern that influences potential to produce flavor and aroma compounds later” (Carvalho et al., 2016). Our results suggest that the production or biosynthetic pathway of some compounds may be more 234 sensitive to early environmental conditions than others. From a crop quality perspective, growers have indicated that their customers would pay more for crops with increased flavor (Walters et al., 2020). However, consumer sensory panels have determined that when consumed raw and alone, there is an upper limit to consider and consumers do not always prefer basil with a more intense flavor (Chapter 4). Therefore, the benefit of an elevated eugenol concentration at harvest is not clear and may be situational. Efficiency implications. In this study, we investigated the effect of increased inputs during the seedling stage to increase yield and secondary metabolite accumulation at harvest. By sowing seeds in a 200-cell tray (1,290 cm2 per flat, 6.45 cm2 per cell), the planting density is 1,550 plants per m2. Seedlings were transplanted 20-cm apart with a planting density of 25 plants per m2. Therefore, planting density was 62 times greater during propagation than finished (harvest) production. Additionally, in this study, the duration of seedling production was 2/3 that of finishing (two weeks compared to three weeks). Taking both the increased planting density and shorter production duration into account, the increase in lighting cost per plant could be discounted by 93 times during seedling versus finished production. Therefore, in this case, the cost per plant of increasing the radiation intensity from 100 to 600 µmol·m‒2·s‒1 during propagation was ~5% that of the cost during finished production. Since the increase in yield at the finishing stage was 80% greater when seedlings were grown under 600 µmol·m-2·s-1 compared to 100 µmol·m-2·s-1 (34.6 compared 5.8 mol·m‒2·d‒1), increasing radiation intensity during propagation increases subsequent yield and eugenol concentration while reducing costs. 235 Conclusions Concentrating resources by increasing radiation intensity from 100 to 600 µmol·m‒2·s‒1 (5.8 to 34.6 mol·m‒2·d‒1) during basil seedling production compared to finished production is a resource-effective method of improving subsequent yields and increasing eugenol concentration. Although elevated CO2 concentrations did not influence fresh or dry mass, future research is needed to determine at what stage of production elevated CO2 concentrations could increase basil growth and secondary metabolite concentrations, if any. With these data, environmental controls, especially radiation intensity and CO2 concentration, can be better leveraged to improve crop productivity, quality, and energy efficiency not only at transplant, but also after finishing in a common environment. 236 APPENDIX 237 Table VI-1. The date of sweet basil ‘Nufar’ (Ocimum basilicum) seed sowing, target and actual CO2 concentration (± SD), target and actual radiation intensity (± SD), and average daily air, canopy, and substrate temperature (± SD) during the seedling growth stage (2 weeks). Rep & start date CO2 (µmol·mol–1) Target Actual 991 ± 19 504 ± 11 503 ± 14 500 1000 500 1000 500 1000 1 7 Aug. 2017 2 10 Nov. 2017 3 22 Jan. 2018 1017 ± 29 z Data not collected 1016 ± 14 506 ± 23 Radiation intensity (µmol·m‒2·s‒1) Target Actual 94 ± 4 200 ± 2 413 ± 6 614 ± 13 102 ± 1 193 ± 4 423 ± 25 589 ± 13 94 ± 4 188 ± 2 432 ± 1 615 ± 2 102 ± 0 191 ± 0 429 ± 1 577 ± 2 88 ± 2 191 ± 3 394 ± 7 589 ± 11 99 ± 2 184 ± 3 384 ± 6 555 ± 9 100 200 400 600 100 200 400 600 100 200 400 600 100 200 400 600 100 200 400 600 100 200 400 600 Temperature (°C) Air 23.0 ± 0.6 23.0 ± 0.1 22.9 ± 1.8 23.0 ± 0.0 23.0 ± 1.4 23.0 ± 0.4 -z Canopy Substrate 25.1 ± 0.6 23.1 ± 0.4 24.4 ± 0.6 23.0 ± 0.5 26.8 ± 1.4 24.1 ± 1.2 27.7 ± 1.8 25.0 ± 1.6 24.2 ± 0.7 23.7 ± 0.8 26.3 ± 1.0 27.1 ± 1.4 24.2 ± 1.2 28.9 ± 1.7 24.7 ± 1.4 24.5 ± 0.9 22.2 ± 1.0 24.0 ± 1.1 22.0 ± 0.9 26.9 ± 1.5 23.7 ± 1.4 27.7 ± 2.1 24.6 ± 1.9 24.4 ± 0.9 21.9 ± 0.7 26.4 ± 1.0 27.4 ± 1.7 24.1 ± 1.3 28.7 ± 1.7 25.2 ± 1.9 24.8 ± 0.9 22.5 ± 0.8 24.1 ± 0.8 22.7 ± 1.0 26.9 ± 1.5 23.5 ± 1.4 27.9 ± 2.2 24.6 ± 1.7 24.2 ± 0.5 21.7 ± 0.6 26.5 ± 0.9 27.4 ± 1.6 24.3 ± 1.5 29.3 ± 2.0 25.3 ± 1.9 - - 238 Table VI-2. Actual average daily air and canopy temperature and daily light integral (DLI) (mean ± SD) during post-transplant greenhouse production (3 weeks of sweet basil ‘Nufar’ (Ocimum basilicum). Rep. 1 2 3 Temperature (°C) Air 21.6 ± 2.1 22.6 ± 1.3 23.1 ± 1.1 Canopy 26.1 ± 3.5 22.6 ± 1.9 22.5 ± 2.4 DLI mol·m‒2·d‒1 14.1 ± 1.8 12.9 ± 3.2 17.4 ± 4.5 239 ) m n · 1 - s · 2 - m · l o m µ ( y t i s n e d x u l f n o t o h p e v i t a l e R 4 3 2 1 0 452 nm 603 nm 662 nm 100 µmol·m-2·s-1 200 µmol·m-2·s-1 400 µmol·m-2·s-1 600 µmol·m-2·s-1 400 500 600 700 800 Figure VI-1. Spectral quality of broad-spectrum light-emitting diode (LED) fixtures providing Wavelength (nm) 20:40:40 blue:green:red radiation ratios (%), a red:far-red ratio of 13:1, and target radiation intensities of 100, 200, 400, or 600 µmol·m–2·s–1. 240 ) m m ( t h g i e H ) 3 m c ( a e r a f a e L ) m m i ( h t d w m e t S 26 24 22 20 18 16 14 0 14 12 10 8 6 4 2 0 2.2 2.0 1.8 1.6 1.4 1.2 1.0 0.0 0 A * y = 1.9E-3+ 6.07E-5x + 6.56E-8x2 R2 = 0.779*** B y = 1.92E1 - 3.25E-2x + 6.77E-5x2 R2 = 0.553*** y = 1.56E1 + 1.00E-2x r2 = 0.335*** * *** y = 5.67E-2 + 8E-4x r2 = 0.772*** D 200 600 Radiation intensity (µmol·m-2·s-1) 400 y = -1.07 + 2.1E-2x r2 = 0.862*** y = -1.51 + 3.04E-2x - 1.81E-5x2 R2 = 0.795*** * C *** y = 9.71E-1 + 1.7E-3x r2 = 0.760*** y = 9.98E-1 + 2.5E-3x - 1.77E-6x2 R2 = 0.694*** * E *** ** 200 600 Radiation intensity (µmol·m-2·s-1) 400 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 0.6 0.5 0.4 0.3 0.2 0.1 0.0 ) g ( s s a m y r D ) g ( s s a m h s e r F Figure VI-2. Radiation intensity (100, 200, 400, or 600 µmol·m–2·s–1) for a 16-h photoperiod to create daily light integrals of (6, 12, 23, or 35 mol·m‒2·d‒1) and CO2 concentration ( , 500 µmol·mol–1; , 1,000 µmol·mol–1; , pooled) effects on sweet basil ‘Nufar’ (Ocimum basilicum) seedling height (A), dry mass (B), leaf area (C), fresh mass (D), and stem width (E) two weeks after sowing. Lines represent linear or quadratic regressions. Symbols (means ± SE) 241 Figure VI-2 (cont’d). represent measured data ( and , n = 30; , n = 60). *, **, and *** indicate significant at P ≤ 0.05, 0.01, or 0.001, respectively. 242 ) m c ( t h g i e H ) m m i ( h t d w m e t S 28 26 24 22 20 18 0 8 7 6 5 0 0 y = 1.84E1 + 1.36E-2x r2 = 0.289*** y = 2.00E1 + 7.7E-3x r2 = 0.134*** y = 5.55 + 3.6E-3x r2 = 0.446*** y = 4.79 + 1.24E-2x - 1.61E-5x2 R2 = 0.232*** * A C * *** 200 600 Radiation intensity (µmol·m-2·s-1) 400 y = 3.74 + 4.3E-3x r2 = 0.161*** y = 3.74 + 5.4E-3x - 4.66E-6x2 R2 = 0.543*** y = 1.61 + 8.7E-3x - 6.71E-6x2 R2 = 0.243*** y = 2.10E1 + 1.14E-1x - 8.62E-5x2 R2 = 0.274*** B D E F 0 200 400 600 Radiation intensity (µmol·m-2·s-1) 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 0.0 5.5 5.0 4.5 4.0 0.0 4.5 4.0 3.5 3.0 2.5 2.0 0.0 70 60 50 40 30 0 ) . o n ( h c n a r B ) . o n ( e d o N ) g ( s s a m y r D ) g ( s s a m h s e r F Figure VI-3. Radiation intensity (100, 200, 400, or 600 µmol·m–2·s–1) for a 16-h photoperiod 243 Figure VI-3 (cont’d). to create daily light integrals of (6, 12, 23, or 35 mol·m‒2·d‒1) and CO2 concentration ( , 500 µmol·mol–1; , 1,000 µmol·mol–1; , pooled) administered during seedling production, two weeks after sowing. The figures depict seedling treatment effects on sweet basil ‘Nufar’ (Ocimum basilicum) height (A), branch number (B), stem width (C), node number (D), dry mass (E), and fresh mass (F) three weeks after transplant into a common enviornment. Lines represent linear or quadratic regressions. Symbols (means ±SE) represent measured data ( and , n = 30; , n = 60). * and *** indicate significant at P ≤ 0.05 or 0.001, respectively. 244 n o i t a r t n e c n o c e l o e n i C 8 1 , ) M D 1 - g m · g n ( n o i t a r t n e c n o c l o o l a n L i ) M D 1 - g m · g n ( n o i t a r t n e c n o c l o n e g u E ) M D 1 - g m · g n ( n o i t a r t n e c n o c l o c i v a h c l y h t e M ) M D 1 - g m · g n ( 1000 800 600 400 0 8000 7000 6000 5000 4000 3000 0 4000 3000 2000 1000 0 6000 4000 2000 0 0 Figure VI-4. Concentrations [ng·mg‒1 dry mass (DM)] of 1,8 cineole (A), linalool (B), eugenol 245 A B C D y = 2.23x2E-2 + -1.31xE1 + 2.85E3 R2 = 0.118** 200 600 Radiation intensity (µmol·m-2·s-1) 400 Figure VI-4 (cont’d). (C), and methyl chavicol (D) of sweet basil ‘Nufar’ (Ocimum basilicum) seedlings grown under radiation intensity (100, 200, 400, or 600 µmol·m–2·s–1) for a 16-h photoperiod to create daily light integrals of (6, 12, 23, or 35 mol·m‒2·d‒1) for two weeks and then transplanted in a common greenhouse environment and grown for three weeks. Each symbol represents the mean of 20 plants ± SE. Lines represent linear or quadratic regression. ** indicates significant at P ≤ 0.01. 246 LITERATURE CITED 247 LITERATURE CITED Ainsworth, E.A. and A. Rogers, 2007. The response of photosynthesis and stomatal conductance to rising [CO2]: Mechanisms and environmental interactions. Plant, Cell, and Environment, 30(3):258−270. Al Jaouni, S., A.M Saleh, M.A. Wadaan, W.N. Hozzein, S. Selim, and H. AbdElgawad, 2018. Elevated CO2 induces a global metabolic change in basil (Ocimum basilicum L.) and peppermint (Mentha piperita L.) and improves their biological activity. Journal of Plant Physiology, 224:121−131. Arena, E., N. Guarrera, S. Campisi, and C.N. Asmundo, 2006. Comparison of odour active compounds detected by gas-chromatography–olfactometry between hand-squeezed juices from different orange varieties. Food Chemistry, 98(1):59–63. Arp, W.J., 1991. Effects of source‐sink relations on photosynthetic acclimation to elevated CO2. Plant, Cell and Environment, 14(8):869−875. Banerjee, C. and L. Adenaeuer, 2014. Up, up and away! The economics of vertical farming. Journal of Agricultural Studies, 2(1):40−60. Beaman, A.R., R.J. Gladon, and J.A. Schrader, 2009. Sweet basil requires an irradiance of 500 µmol·m2·s-1 for greatest edible biomass production. HortScience. 44(1):64−67. Berry, J.A. and J.K. Raison, 1981. Responses of macrophytes to temperature. In Physiological Plant Ecology I (pp. 277−338). Springer, Berlin, Heidelberg. Besford, R.T., L.J. Ludwig, and A.C. Withers, 1990. The greenhouse effect: Acclimation of tomato plants growing in high CO2, photosynthesis and ribulose-1, 5-bis phosphate carboxylase protein. Journal of Experimental Botany, 41(8):925−931. Blackman, F.F., 1905. Optima and limiting factors. Annals of Botany, 19(74):281−295. Carvalho, S.D., M.L. Schwieterman, C.E. Abrahan, T.A. Colquhoun, and K.M. Folta, 2016. Light quality dependent changes in morphology, antioxidant capacity, and volatile production in sweet basil (Ocimum basilicum). Frontiers in Plant Science, 7:1328. Chang, X., P. Alderson, and C. Wright, 2005. Effect of temperature integration on the growth and volatile oil content of basil (Ocimum basilicum L.). The Journal of Horticultural Science and Biotechnology, 80:593‒598. Chang, X., P.G. Alderson, and C.J. Wright, 2008. Solar irradiance level alters the growth of basil (Ocimum basilicum L.) and its content of volatile oils. Environmental and Experimental Botany, 63(1):216−223. 248 Chermnykh, L.N. and A.A. Kosobrukhov, 1987. Effect of environmental factors on optimum temperature and photosynthetic intensity of plants adapted to various conditions. Biotronics, 16:1‒11. Craver, J.K., 2018. Manipulating light quality, light intensity, and carbon dioxide concentration to optimize indoor and greenhouse production of annual bedding plant seedlings. Purdue University, West Lafayette, IN, PhD Dissertation De Vincenzi, M., M. Silano, A. De Vincenzi, F. Maialetti, and B. Scazzocchio, 2002. Constituents of aromatic plants: Eucalyptol. Fitoterapia, 73(3):269–275. Dou, H., G. Niu, M. Gu, and J.G. Masabni, 2018. Responses of sweet basil to different daily light integrals in photosynthesis, morphology, yield, and nutritional quality. HortScience, 53(4):496–503. Gibson, K.E., A.J. Lamm, F. Masambuka-Kanchewa, P.R. Fisher, and C. Gómez, 2020. Identifying indoor plant propagation research and education needs of specialty crop growers. HortTechnology, 1(aop):1-9. Goodman, W. and J. Minner, 2019. Will the urban agricultural revolution be vertical and soilless? A case study of controlled environment agriculture in New York City. Land Use Policy, 83:160–173. Hurd, R.G. and J.H.M. Thornley, 1974. An analysis of the growth of young tomato plants in water culture at different light integrals and CO2 concentrations: I. Physiological aspects. Annals of Botany, 38(2):375−388. Korczynski, P.C., J. Logan, and J.E. Faust, 2002. Mapping monthly distribution of daily light integrals across the contiguous United States. HortTechnology, 12(1):12–16. Kozai, T., 2013. Resource use efficiency of closed plant production system with artificial light: Concept, estimation and application to plant factory. Proceedings of the Japan Academy, Series B, 89:447−461. Krizek, D.T., W.A. Bailey, H. Klueter, and R.C. Liu, 1972. Maximizing growth of vegetable seedlings in controlled environments at elevated temperature, light and CO2. Acta Horticulturae, 39:89−102. Lindhout, P. and G. Pet, 1990. Effects of CO2 enrichment on young plant growth of 96 genotypes of tomato (Lycopersicon esculentum). Euphytica, 51(2):191−196. Litvin, A.G., 2019. Quantifying the effects of light quantity and quality on culinary herb physiology. Iowa State University, Ames, PhD Dissertation. Long, S.P., 1991. Modification of the response of photosynthetic productivity to rising temperature by atmospheric CO2 concentrations: Has its importance been underestimated? Plant, Cell, and Environment, 14(8):729−739. 249 Ogren, W.L., and G. Bowes. 1971. Ribulose diphosphate carboxylase regulates soybean photorespiration. Nature New Biology, 230(13):159−160. Park, K.S., K. Bekhzod, J.K. Kwon, and J.E. Son, 2016. Development of a coupled photosynthetic model of sweet basil hydroponically grown in plant factories. Horticulture, Environment, and Biotechnology, 57(1):20−26. Pramuk, L.A. and E.S. Runkle, 2005. Photosynthetic daily light integral during the seedling stage influences subsequent growth and flowering of Celosia, Impatiens, Salvia, Tagetes, and Viola. HortScience, 40(4):1099C−1099. Sage, R.F., T.D. Sharkey, and J.R. Seemann, 1989. Acclimation of photosynthesis to elevated CO2 in five C3 species. Plant Physiology, 89(2):590−596. Santos, A., G. Chierice, K. Alexander, A. Riga, and E. Matthews, 2009. Characterization of the raw essential oil eugenol extracted from Syzygium aromaticum L. Journal of Thermal Analysis and Calorimetry, 96(3):821–825. Schilmiller, A., F. Shi, J. Kim, A.L. Charbonneau, D. Holmes, A.D. Jones, and R.L. Last, 2010. Mass spectrometry screening reveals widespread diversity in trichome specialized metabolites of tomato chromosomal substitution lines. The Plant Journal, 62(3):391–403. Sharkey, T.D., 1985. Photosynthesis in intact leaves of C3 plants: Physics, physiology and rate limitations. The Botanical Review, 51(1):53–105. Simon, J.E., M.R. Morales, W.B. Phippen, R.F. Vieira, and Z. Hao, 1999. Basil: A source of aroma compounds and a popular culinary and ornamental herb. p. 499–505. In: J. Janick (ed.). Perspectives on new crops and new uses. ASHS Press. Arlington, VA. Thomas, J.F., C.D. Raper Jr., C.E. Anderson, and R.J. Downs, 1975. Growth of young tobacco plants as affected by carbon dioxide and nutrient variables. Agronomy Journal, 67(5):685–689. Tremblay, N. and A. Gosselin, 1998. Effect of carbon dioxide enrichment and light. HortTechnology, 8:524–528. U.S. Agency for International Development-Acceso. 2014. The U.S. market for fresh herbs. Market Brief #20. http://pdf.usaid.gov/pdf_docs/PA00KP2C.pdf Accessed 23 Apr. 2019. U.S. Department of Agriculture – National Agriculture Statistics Service (USDA-NASS). 2015. 2014 Census of Horticulture Specialties. https://www.nass.usda.gov/Publications/AgCensus/2012/Online_Resources/Census_of_H orticulture_Specialties/ Accessed 23 Apr. 2019. Walters, K.J. and C.J. Currey, 2015. Hydroponic greenhouse basil production: Comparing systems and cultivars. HortTechnology. 25(5):645−650. 250 Walters, K.J. and C.J. Currey, 2018. Effects of nutrient solution concentration and daily light integral on growth and nutrient concentration of several basil species in hydroponic production. HortScience, 53(9):1319−1325. Walters, K.J., B.K. Behe, C.J. Currey, and R.G. Lopez. 2020. Historical, current, and future perspectives for controlled environment hydroponic food crop production in the United States. HortScience, 55(6):758−767. Wang, S.Y., and J.A. Bunce, 2004. Elevated carbon dioxide affects fruit flavor in field‐grown strawberries (Fragaria× ananassa Duch). Journal of the Science of Food and Agriculture, 84(12):1464−1468. Wittwer, S.H. and W.M. Robb, 1964. Carbon dioxide enrichment of greenhouse atmospheres for food crop production. Economic Botany, 18(1):34−56. 251