This study was conducted in a small plastic greenhouse (4.0 × 3.0 × 3.0 m) with a three-layer polyethylene (PE) film structure, located at the affiliated farm of Mokpo National University in Muan-gun, Jeollanam-do (Fig. 1A). The experiment took place from February 21 to June 8, 2023, using the "Dafnis" cultivar (Syngenta, Switzerland). The plants used in the experiment were first sown in germination trays with 240 rockwool plugs and then transplanted to coir slabs (110 × 20 × 10 cm, Pelemix, Israel) when 2–3 true leaves had emerged. The nutrient solution was supplied at an EC range of 2.0–3.0 dS · m⁻¹ and a pH range of 5.5–6.5, depending on the growth stage. For the supply cycle, 100–180 mL of solution was applied for every 80 J · cm⁻² of accumulated solar radiation, depending on the growth stage. In addition, to prevent a rapid drop in temperature at night, an electric heater (SWE-15F, Shinwoo Automatic, Korea) was used and set to activate when the greenhouse temperature dropped to 16°C or below. The ventilation window was set to open and close when the greenhouse temperature reached 26°C or higher, allowing ventilation as the temperature increased. Fertilization was set to stop when the ventilation window opened and closed by 40% or more.

To determine the effects of supplemental lighting and CO₂ enrichment on tomatoes in a plastic greenhouse, the experimental treatments were divided into four groups: control without both supplemental lighting and CO₂ enrichment (CON); CO₂ enrichment treatment (CE); supplemental lighting treatment (SL); and combined treatment (CT). CO₂ enrichment, which can increase the photosynthetic rate, used liquefied carbon dioxide for food and beverage use (Fig. 1B) and was set at 600 μmol-mol⁻¹. CO₂ enrichment was applied from 6:00 am to 11:00 am, and data were recorded using a CO₂ sensor (Telaire T 7001, Telaire, CA, USA) inside the greenhouse. Supplemental lighting was provided based on the daily accumulated solar radiation. A radiation sensor (SP-215-SS, Apogee Instruments, UT, USA) and a photosynthetic photon flux density (PPFD) sensor (SQ-212, Apogee Instruments, UT, USA) were installed outside the greenhouse, while another PPFD sensor (SQ-212, Apogee Instruments, UT, USA) was used to measure the PPFD inside the greenhouse. Supplemental lighting and CO₂ treatment were initiated 30 days after planting. Supplemental LED lighting was operated from 06:00 to 22:00 and was activated when solar radiation outside the greenhouse decreased to 100 W · m⁻² or less. The light source for the supplemental lighting treatment was a long LED bar (Allix, Korea; Fig. 1C), which was maintained at an intensity of 100 μmol · m⁻² · s⁻¹ at a point 1 m vertically above the light source (Table 1).

Photosynthesis of individual tomato leaves was measured using a portable photosynthesis meter (LI-6800, LI-COR, NE, USA). The settings in the chamber of the photosynthesis meter were as follows: flow rate of 600 μmol · mol⁻¹, leaf temperature of 25°C, relative humidity (RH) of 55%, and PPFD of 1,500 μmol · m⁻² · s⁻¹. During the measurement of the photosynthetic rate in response to CO₂, the CO₂ concentration was changed stepwise. The photosynthetic response to CO₂ was measured by first reducing the CO₂ concentration from 400 μmol · mol⁻¹ to 50 μmol · mol⁻¹ to remove residual CO₂ in the mesophyll, followed by a gradual increase to 1,600 μmol · mol⁻¹. For the light response measurements, the CO₂ concentration was gradually increased from 0 μmol · mol⁻¹ to 1,500 μmol · mol⁻¹. Using this method, the photosynthetic response was analyzed at each concentration, and the measurements were taken from a single leaf located at the third leaf position, based on the inflorescences.

Equation 1 describes how the photosynthetic rate (A) increases with CO₂ concentration (I). By estimating the maximum photosynthetic rate (A_max) and day respiration rate (R_d), the photosynthetic rate of tomatoes can be predicted based on CO₂ concentrations.

The FvCB model was used to analyze the photo-synthetic rate as a function of CO₂ concentration in the mesophyll. The photosynthetic rate (A_net, μmol CO₂ · m⁻² · s⁻¹; Eq. 2) in the FvCB model was estimated using the maximum reaction rate of the enzyme RuBisCO (A_c, μmol CO₂ · m⁻² · s⁻¹; Eq. 3), the electron transport limitation (A_j, μmol CO₂ · m⁻² · s⁻¹; Eq. 4), and the day respiration rate (R_d, μmol CO₂ · m⁻² · s⁻¹). A_c is expressed as a function of the maximum carboxylation rate (V_cmax, μmol CO₂ · m⁻² · s⁻¹), while A_j is expressed as a function of the maximum electron transport rate (J_max, μmol CO₂ · m⁻² · s⁻¹)

To compare tomato growth across different experimental treatments, the first growth survey was conducted 31 days after planting. Subsequent surveys were carried out every two weeks, with 8 plants per treatment. Survey items included plant height, stem diameter, number of leaves, leaf area, fresh weight, dry weight, chlorophyll index, chlorophyll fluorescence, and fruit characteristics. Plant height was measured from the stem base at the soil surface to the growing point, while stem diameter was measured just below the inflorescences using a vernier caliper (DC200-2, CAS, Korea). Leaf number was recorded as the count of fully expanded leaves below the inflorescences. Leaf area was measured using the ImageJ program (v1.8.0, National Institutes of Health, MD, USA). Fresh weight was recorded with an electronic balance (IB-3100, InnoTem, Korea), while dry weight was measured using the same balance after drying the plants in an oven (JSOF-250T, JSR, Korea) at 70°C for 72 hours. Soil Plant Analysis Development (SPAD), a relative chlorophyll content index, was measured using a portable SPAD meter (SPAD-502, Konica Minolta Inc., Japan). The ratio of variable to maximum fluorescence (Fv/Fm) was measured using a portable chlorophyll fluorometer (FluorPen FP 110/D, Photon Systems Instruments, Czech Republic). Yield was estimated by converting the fruit weight and planting density to 10a for plants with fruits per inflorescence.

Environmental data from the greenhouse were collected using a data logger (CR1000X, Campbell Scientific, UT, USA). Graphs were generated with SigmaPlot software (SigmaPlot 14.5, Systat Software Inc., CA, USA), and tomato growth data were analyzed using SPSS statistical software (IBM SPSS Statistics, IBM, USA).

Fig. 2 showed graphs of the outside PPFD, as well as the temperature and relative humidity inside the greenhouse throughout the experimental period. As the experiment progressed, the outside PPFD exhibited a steady upward trend, although occasional decreases were observed due to cloudy weather and rain. The temperature inside the greenhouse was maintained at an average of 28°C throughout the experimental period. However, as the period shifted to a high-temperature phase, the maximum temperature rose to approximately 37°C. Moreover, to prevent a drop in nighttime temperatures, an electric heater was set to activate at 16°C, ensuring stable nighttime temperatures inside the greenhouse. The relative humidity inside the greenhouse ranged from 30% to 90%, controlled by adjusting the ventilation in response to the internal temperature. The graphs in Fig. 3 show the changes in CO₂ concentration and inside PPFD measured over a 7-day period following the experimental treatments. In Fig. 3A, the CO₂ concentration in both the control (CON) group and the supplemental lighting (SL) group, without CO₂ enrichment, was maintained between 400 and 450 μmol · mol⁻¹, similar to the average atmospheric concentration. In contrast, the CO₂ enrichment (CE) group and the combined treatment (CT) group, which received both CO₂ enrichment and supplemental lighting, achieved the target CO₂ concentration of 600 μmol · mol⁻¹. However, this concentration gradually decreased over time due to CO₂ uptake by the tomatoes and the operation of the ventilation window in response to rising temperatures inside the greenhouse. Since photosynthesis is most active after sunrise, the fertilization setting was designed to maximize the photosynthetic rate by increasing the CO₂ concentration in the greenhouse through CO₂ enrichment before sunrise. However, as the experiment progressed and temperatures rose during the high-temperature period, the side ventilation windows were activated to open and close in response to the temperature increase, resulting in uneven CO₂ enrichment.

The graph showing the changes in light intensity inside the greenhouse presents the results from measurements taken with a PPFD sensor. As shown in Fig. 3B, the average daily inside PPFD was 53.6 μmol · m⁻² · s⁻¹ for the CON group and 69.7 μmol · m⁻² · s⁻¹ for the CE group. Both the CON and CE groups displayed similar patterns of daily PPFD changes, which can be attributed to their reliance solely on natural light. On the other hand, the average daily inside PPFD was 113.9 μmol · m⁻² · s⁻¹ for the SL group and 121.8 μmol · m⁻² · s⁻¹ for the CT group, which had higher PPFD values because the supplemental LED lighting was activated when the target average daily PPFD was insufficient. It appears that the light deficit, caused by factors such as cloudy weather, the geographical location surrounded by mountains, and reduced light transmittance from long-term use of covering materials, was addressed by the supplemental LED lighting.

Fig. 4A and Table 2 present a comparison of the photosynthetic rates of tomato leaves across treatment groups, based on changes in CO₂ concentration within the leaves. A_max, the maximum photosynthetic rate, for each treatment group was 26.74, 31.32, 28.15, and 29.10 μmol · m⁻² · s⁻¹ for the CON, CE, SL, and CT groups, respectively. The CON group exhibited a generally lower photosynthetic rate compared to the other treatment groups, while the CE group showed a rapid increase in photosynthetic rate when CO₂ concentration reached 400 μmol · mol⁻¹, maintaining a high rate at elevated CO₂ concentrations. The SL and CT groups exhibited higher photosynthetic rates than the CON and CE groups at low CO₂ concentrations, but their rates were lower than the CE group as CO₂ concentrations increased. It appears that the increased photosynthetic rate in the CE group was due to enhanced photosynthesis, driven by the increased CO₂ partial pressure in the plant leaves at higher CO₂ concentrations (Kim and Lee, 2001).

Regarding the photosynthetic rate of tomatoes based on light intensity (Fig. 4B, Table 3), the A_max values were 12.91, 13.77, 15.04, and 17.08 μmol · m⁻² · s⁻¹ for the CON, CE, SL, and CT groups, respectively. These results suggest that tomatoes treated with supplemental lighting have a greater adaptability to light intensity than those without supplemental lighting. The higher A_max value observed in the CT group, compared to the other treatment groups, is likely due to the combined application of CO₂ enrichment and supplemental lighting, which had a positive effect on photosynthesis; it also appears that supplemental lighting improved the photosynthetic rate under low light conditions and promoted RuBisCO activity and RuBP regeneration, thereby increasing the overall photosynthetic rate (Lu et al., 2012).

An analysis of the photosynthetic response curve to CO₂ in the mesophyll, using the FvCB model, found element-specific limitations based on CO₂ concentration (Fig. 5, Table 4). V_cmax, the maximum carboxylation rate, was 66.98, 66.12, 69.11, and 82.22 μmol · m⁻² · s⁻¹ for the CON, CE, SL, and CT groups, respectively. J_max, the maximum electron transport rate, was 138.38, 145.28, 142.57, and 146.09 μmol · m⁻² · s⁻¹ for the CON, CE, SL, and CT groups, respectively. In the FvCB model, no significant differences were found in V_cmax, J_max, and R_d among the treatment groups, except for the V_cmax of the CT group. The CT group exhibited higher V_cmax and J_max values by 15.24 μmol · m⁻² · s⁻¹ and 7.71 μ mol · m⁻² · s⁻¹, respectively, compared to the CON group. Given the large difference in V_cmax and the smaller difference in J_max, it appears that the CT group experienced greater limitations in RuBisCO activity and fewer limitations in RuBP regeneration compared to the CE and SL groups (Duursma, 2015).

After planting, plant height measurements showed a consistent increase in all treatment groups (Fig. 6). However, on day 93 after planting, plant height was higher in the CT group than in the other groups. This could be due to greater nutrient allocation to the fruit in the CT group, as tomato is a crop that undergoes both vegetative and reproductive growth. Stem diameter initially increased rapidly in all treatment groups up to day 60, but then showed a tendency to decrease. This suggests weak initial growth, especially when compared to the findings of a previous study (Stradiot and Battistel, 2002), which reported that a stem diameter of 11–12 mm or greater is a sign of strong growth.

Throughout the experimental period, the number of leaves increased steadily in all treatment groups. To prevent a decrease in the photosynthetic rate and ensure adequate ventilation, old leaves were regularly defoliated. Leaf area index (LAI), which indicates the leaf area per unit ground area, was initially higher in the CON group than in the SL group. However, as growth progressed, the LAI in the CON group decreased rapidly, notably dropping below 1.0 in the later stages. This decrease is likely due to the weak initial growth observed. The treatment group with the highest rate of increase in LAI during the growth period was the CE group, suggesting that CO₂ enrichment promoted greater growth through an increase in leaf area. The SL group showed minimal changes in LAI during the early stages of growth, but the LAI gradually increased in the later stages of growth as growth progressed. This suggests that supplemental lighting has a positive effect on increasing leaf area in the later stages of growth. Fresh weight was highest in the CT group until the middle of the experiment, but the CE group recorded the highest fresh weight in the latter part of the experiment. The greatest difference in dry weight was observed in the CE group, followed by the CT, CON, and SL groups. The CON and SL groups exhibited similar decreases in dry weight during the final growth survey. These results indicate the long-term effects of CO₂ enrichment on plant dry weight and are consistent with a previous study showing its effects on plant biomass accumulation during the growth period (Heuvelink, 1999).

The average SPAD (relative chlorophyll content index) for tomatoes ranges from 40 to 60 (Jiang et al., 2017), and all treatment groups in this study fell within this range. However, the treatment groups showed higher SPAD values compared to the CON group during the experimental period. Specifically, 63 days after the experimental treatment, the SPAD values for the CON, CE, SL, and CT groups were 51.9, 53.1, 55.6, and 56.1, respectively (Fig. 7A). Fv/Fm is a fluorescence parameter that indicates the maximum quantum efficiency of photosystem II (PSII). It is known to decrease when plants are exposed to stress or when electron transport is inhibited (Bjorkman et al., 1987; Choi, 2020). On day 21 after the experimental treatments, the Fv/Fm values for the CON, CE, SL, and CT groups were 0.75, 0.76, 0.78, and 0.78, respectively. On day 42, the values for these groups were 0.79, 0.79, 0.80, and 0.80, respectively. By day 63, the Fv/Fm values for the CON, CE, SL, and CT groups were 0.78, 0.80, 0.80, and 0.80, respectively (Fig. 7B).

Yield data collected 7 weeks after planting the tomato seedlings are presented in Fig. 8, which compares the actual yields for each experimental treatment. Initial yields in all treatment groups showed similar trends. However, over time, the yields in the other treatment groups tended to be higher than those in the CON group. Likely due to the experimental site being surrounded by mountains, which resulted in insufficient daylight hours, and the long-term use of covering materials that decreased light transmittance, it appears that the CT group, which received both CO₂ enrichment and supplemental light treatments, had the highest yield. Tomato yields in the CE and SL groups were higher than those in the CON group due to their individual treatments, but still lower than those in the CT group. In the CE group, the photosynthetic rate increased with higher CO₂ concentrations; however, the total photosynthetic rate reached a limit due to insufficient light intensity (PPFD) to fully assimilate the increased CO₂. On the other hand, the SL group increased the energy supply by providing sufficient light intensity, but the CO₂ concentration required for glucose production reached a limit. The highest tomato yield was observed in the CT group, with the CE, SL and CT groups yielding 15%, 22% and 29% more than the CON group, respectively. These results suggest that the combined treatment of supplemental lighting and CO₂ enrichment had a synergistic effect on tomato growth and yield (Lu et al., 2012; An et al., 2011). The high yield of the CT group indicates that tomato growth was improved when both CO₂ and light intensity were sufficiently provided simultaneously, which played an important role in overcoming the environmental limitations of the experimental site. In conclusion, this study supports the effectiveness of the combined treatment of supplemental lighting and CO₂ enrichment in improving tomato growth and yield. Based on this, the key environmental conditions were determined that can maximize productivity by simultaneously optimizing the main factors affecting photosynthesis in tomato cultivation. These findings are expected to contribute to increasing tomato productivity in Korea through CO₂ fertilization and supplemental lighting treatments in greenhouse tomato production in the future.