Li 1400

Photosynthetic active radiation (PAR) sensor (Li-190SA, Li-COR, USA) was installed above the plants in the field to collect light intensity, and was connected to a data logger (Li-1400, Li-COR, USA). This enabled us to measure the PAR value, its maximum, and to calculate the total input and to obtain average values of PAR for each treatment during canopy development. The total PAR input of any leaf was calculated as a sum of incident PAR (in mols of photons per unit area per second) between the appearance of the leaf and the time of performing photosynthesis and fluorescence measurements and the HL treatment. The middle part of mature leaves of barley (which was measured) was almost in a horizontal position; hence, the measured values of PAR almost fully corresponded to light intensities incident on leaves.

Physicochemical profiles (depth, temperature, salinity) of the water column were obtained with a multi-parameter probe (Hydrolab Data sonde 5 Options, USA), and vertical profiles of irradiance were measured with an underwater quantum sensor (LICOR LI-1400, Nebraska, USA). Inorganic nutrients analysis was performed in the laboratory with a Bran+Luebbe Autoanalyzer AA3 according to Aminot & Kérouel [21] . Chlorophyll a biomass (Chl a) was estimated by fluorometry (Trilogy 7200-000 - Turner Designs, California, USA) according to the method of Welschmeyer [22] as described in Bazin et al. [23] .

UVB exposure was provided with UVB-fluorescent tubes (FL20SE; Toshiba or GL20SE; Sankyo Denki). Strong light exposure was provided by a xenon light source (MAX-303; Asahi Spectra) equipped with a mirror module (MAX-VIS; Asahi Spectra) to extract visible light (wavelength 385–740 nm) and a rod lens (RLQL80-1; Asahi Spectra) to emit light with uniform intensity in a chamber. The intensity of UVB or visible light was measured with a data logger (LI-1400; Li-Cor) equipped with a UVB sensor (SD204B; Li-Cor) or a photosynthetic photon flux density sensor (LI-190SA; Li-Cor), respectively. After the treatment, plants were cultivated in the indicated growth conditions until the analysis.
For ConA treatment to suppress lytic activity in the vacuole, MES-NaOH (pH 5.5) containing 1 �M ConA was infiltrated immediately after UVB treatment into leaves with a 1-ml syringe and the leaves incubated under the indicated growth conditions. After 1 or 2 d, leaf mesophyll cells were observed under the confocal microscopy.

Qualitative analysis of flavor components: Through computer search and comparison with standard mass spectra provided by NIST 105 and Wiley 7.0 mass spectral libraries, the volatile components of tomato identified by GCMS analysis were analyzed to determine their chemical components.
Quantitative analysis of flavor components: The NIST spectral library workstation data processing system was used for quantitative analysis according to the peak area normalization method, and the percentage content of each chemical component in the volatile components of tomato was obtained.
The calculation method of daily light integral (DLI):
where DLI is the daily cumulative light amount, mol/m²/d; the unit of light intensity is μmol/m²/s; and the unit of photoperiod is h/d [8 (link)].
Light intensity and spectrum measurement: light intensity was measured at 20 cm directly below the LED lamps using a portable light quantum meter (LI-1400, LI-COR, Lincoln, NE, USA), and the spectral distributions of light were measured by a spectrometer (AvaSpec-ULS2048, Avantes, NS Apeldoorn, The Netherlands) (Figure 3). According to the spectral distribution, the ratios of blue light (B, wavelength 400–499 nm), green light (G, wavelength 500–599 nm), and red light (R, wavelength 600–699 nm) were calculated, respectively.

Light interception (LI) data were collected from 10.00 to 10.30 am every 15 days during the growing season at 30, 45, 60, 75, 90, 105, and 120 days of emergence (DAE) on days when there were no clouds. The spatial-grid sampling method was used to measure the transmitted Photosynthetically Active Radiation (tPAR) and the reflected radiation in multiple layers of the canopy with the help of a movable 1.0 m-line light sensor (LI-191SA,LI-COR,Lincoln,NE,USA) and a datalogger (LI-1400, LI-COR). The TPAR and RPAR were sampled starting from a row on the west side to a neighboring east row, and in five sections in each plot with horizontal distances of 0 cm, 20 cm, 40 cm, 60 cm, and 80 cm. Similarly, the whole cotton canopy was separated into consecutive strata from the lowermost strata then every 20 cm to the uppermost strata of the canopy, with several strata fluctuating in plant height. The following equations from Bai [10 (link)] were used to calculate a significant portion of the transmitted Photosynthetically Active Radiation (tPAR), the reflected portion of radiation, and the intercepted amount of Photosynthetically Active Radiation (iPAR): $$t PAR={\rm T}{\rm P}AR{\rm I}{\rm P}AR$$
$$r {\rm P}AR=R{\rm P}AR{\rm I}{\rm P}AR$$
$$i {\rm P}AR=I PAR-TPAR-RPARIPAR=1-t PAR-rPAR$$
The effect of rPAR is ignored in this study. Equation (3) is simplified as follows: $$iPAR=1-tPAR$$

To characterize the effect of the water optical properties on light availability across depths, we measured the diffuse attenuation coefficient for downwelling irradiance (K_d) at the beginning of the experiment. K_d was calculated by measuring changes in light intensities across the depth gradient using the cosine-corrected photosynthetically active radiation (PAR) sensor of a diving-pulse amplitude modulated (PAM) (Walz), previously calibrated against a manufacturer-calibrated quantum sensor (LI-1400, LI-COR). The available light intensity at the depth of each transplant site, expressed as the percentage of incident light, was calculated based on the local K_d as E_z = E₀ e^−Kdz [25 ], where E_z is the % irradiance at z depth (in metres) and E₀ is the % irradiance at sea surface (100%). Variation in temperature and relative light levels throughout the duration of the experiment was recorded every 30 min from 26 September 2014 until 20 March 2015 by Onset HOBO data loggers (UA-002-64, Onset Computer Corporation) attached to the PVC panels (one logger per panel).

Minimum realized temperature during the diel cycle increased gradually from 7°C in February to 12°C in June and July, maximum temperature increased from 17 to 30°C. Liquid CO₂ was used to enrich the greenhouse air to 600–800 ppm when vents were closed. During ventilation, greenhouse air was kept at ambient CO₂ level.
Solar radiation was recorded every 5 min. based on a Kipp solarimeter placed outside the glasshouse. Three quantum sensors (Li-190R, Li-Cor Inc., Lincoln, NE, United States) were placed inside each glasshouse compartment, 3.50 m above floor level, near the top of the glasshouse, to measure incoming photosynthetically active radiation (PAR). These sensors were connected to a data logger (Li-1400, Li-Cor Inc., Lincoln, NE, United States). Fraction PAR in solar radiation was assumed to be 0.5 (Jacovides et al., 2004 (link)). Greenhouse transmissivity was calculated as the ratio between measured PAR inside the greenhouse and calculated PAR outside.
The fruiting laterals were trellised according to commercial standards. At the onset of flowering, a small hive of bumblebees was introduced in the greenhouse compartments. Two weeks later, the bumblebees were removed and replaced by honey bees. A three stage (vegetative growth, flowering and fruiting) standard blackberry nutrient solution was applied according to commercial standards.