*3.4. Plant Material and Growth Conditions*

Seeds of *Thymus vulgaris* L. (Rühlemann's Kräuter- und Duftpflanzen, Horstedt, Germany) were sown in 128-cell plug trays (Ø 4 cm) filled with potting substrate (Fruhstorfer Einheitserde Typ P, HAWITA, Vechta, Germany) on 9 October 2018 and placed under the differing lighting fixtures in a NS-orientated greenhouse located in Berlin, Germany (52.5◦ N, 13.3◦ E). After six weeks on 20 November 2018, 32 representative seedlings were transplanted into pots (Ø 9 cm) containing substrate with an elevated nutrient composition (Fruhstorfer Einheitserde Typ T, HAWITA, Vechta, Germany) and evenly placed (quadratic) within 1 m<sup>2</sup> under each light fixture. Each treatment was replicated four times, positioned in a randomized block design, and surrounded by 28 border plants to avoid boundary effects. Starting 27 December 2018, plants were fertilized weekly with 100 mL of a 0.2% (*v*/*v*) nutrient solution (Hakaphos® Blau, COMPO EXPERT®, Muenster, Germany) containing 15% N, 10% P2O5, 15% K2O, 2% MgO, 0.01% B, 0.02% Cu, 0.075% Fe, 0.05% Mn, 0.001% Mo, and 0.015% Zn. However, due to the low biomass accumulation of the plants grown under the fluorescent lamp system, the plant fertilization was started two weeks later under

the FL treatment. All thyme plants grew for a total period of 18 weeks until harvest on 12 February 2019. Climatic conditions including temperature and relative humidity of the greenhouse air were continuously monitored at canopy level via data loggers (EL-USB−2, Lascar, CONRAD, Hirschau, Germany). Average temperatures ( ◦C ± SD) under LED, HPS, and FL lighting were 20.4 ± 1.4, 21.1 ± 2.0, and 20.9 ± 1.2, with a measuring accuracy of 1 ◦C. Average humidities (%rh ± SD) under LED, HPS, and FL lighting were 47.2 ± 7.0, 44.2 ± 7.1, and 38.8 ± 7.1 with a measuring accuracy of 2.25%rh. Both climatic conditions did not differ between treatments. spectra and detailed spectral compositions of the supplemental lighting systems are depicted in Figure 4 and summarized in Table 1, respectively. Under each lighting system, the irradiance profiles within one square meter (representing the replicated experimental plots during the greenhouse experiment) were measured over a flat plane below the fixtures at intervals of 10 cm between 360 and 760 nm. Each measurement represents the spectral irradiance in W m−2 nm−1 and was replicated three times and averaged, leading to a total dataset of 100 measurements per square meter. These irradiance profiles are depicted in Figure 3.

spectrometer (uSpectrum PC laboratory software) automatically calculated all electro-

irradiance (W m−2 nm−1) between 360 and 760 nm and photosynthetic photon flux density

between 350 and 800 nm of the natural irradiance through the greenhouse glass was determined to be 28% (± 5%) by comparing the output of the spectrometer inside the greenhouse with the output outside the greenhouse and resulted in a photon flux density of ~

tegral of 3.9 mol m−2 d−1 when combined with the weather recordings (Section 3.2). Light

−1) between 400 and 700 nm. During a sunny day, light transmission

−1 at bench level, which amounts to an approximate natural daily light in-

−1) and spectral

magnetic parameters including photon flux density (PFD in µmol m−2 s

*Plants* **2021**, *10*, x FOR PEER REVIEW 11 of 17

(PPFD in µmol m−2 s

434 µmol m−2 s

**Figure 4.** Light spectra of the three artificial light sources (light-emitting diode (LED) = solid line, high-pressure sodium lamp (HPS) = dashed line, fluorescent light (FL) = dotted line) used during the greenhouse experiment. **Figure 4.** Light spectra of the three artificial light sources (light-emitting diode (LED) = solid line, high-pressure sodium lamp (HPS) = dashed line, fluorescent light (FL) = dotted line) used during the greenhouse experiment.

#### *3.4. Plant Material and Growth Conditions 3.5. Harvest and Crop Managements*

Seeds of *Thymus vulgaris* L. (Rühlemann`s Kräuter- und Duftpflanzen, Horstedt, Germany) were sown in 128-cell plug trays (Ø 4 cm) filled with potting substrate (Fruhstorfer Einheitserde Typ P, HAWITA, Vechta, Germany) on 9 October 2018 and placed under the differing lighting fixtures in a NS-orientated greenhouse located in Berlin, Germany (52.5° N, 13.3°E). After six weeks on 20 November 2018, 32 representative seedlings were transplanted into pots (Ø 9 cm) containing substrate with an elevated nutrient composition (Fruhstorfer Einheitserde Typ T, HAWITA, Vechta, Germany) and evenly placed (quadratic) within 1 m<sup>2</sup> under each light fixture. Each treatment was replicated four times, positioned in a randomized block design, and surrounded by 28 border plants to avoid To analyze the effect of the supplemental lighting systems on the yield, all 32 experimental thyme plants were harvested separately from each treatment condition and replication. Fresh matter (FM) of the above-ground plant parts was individually recorded at harvest on 12 February 2019. Total dry matter (DM) was measured after drying the samples in a circulated drying oven at 30 ◦C until stable mass was attained (≤seven days). Leaf dry matter (LDM) was determined for 16 plants selected from each treatment and replication, and the corresponding shoot dry matter (SDM) was calculated by subtracting the LDM from DM. All dried leaf samples were vacuum-sealed (V.300®, Landig + Lava GmbH & Co KG, Bad Saulgau, Germany) and stored in the dark at 4 ◦C until further processing.

#### boundary effects. Starting 27 December 2018, plants were fertilized weekly with 100 mL of a 0.2% (v/v) nutrient solution (Hakaphos® Blau, COMPO EXPERT® *3.6. Energy Measurements*

, Muenster, Germany) containing 15% N, 10% P2O5, 15% K2O, 2% MgO, 0.01% B, 0.02% Cu, 0.075% Fe, 0.05% Mn, 0.001% Mo, and 0.015% Zn. However, due to the low biomass accumulation of the plants grown under the fluorescent lamp system, the plant fertilization was started The power draw of current (I) and voltage (U) as well as electrical characteristics including real power (P) and apparent power (S) from representative lamps of each lighting treatment were measured using a power meter (ENERGY MONITOR 3000, VOLTCRAFT®, Wernberg-Köblitz, Germany) in order to estimate energy consumptions and biomass efficacies of the light fixtures. To correct for the detected difference between P and S due to heat dissipation of the HPS system, the measured *cos phi* of 0.93 was incorporated into the HPS' power consumption calculations. According to the manufacturer's specifications, the HPS system allows a homogeneously illuminated area of 1.56 m<sup>2</sup> (1.2 × 1.3 m). Thus, the measured power consumptions were adjusted to the power consumption per square meter (W m−<sup>2</sup> ) via rule of three. No adjustments were necessary for the LED and FL system.
